| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Ann Thorac Surg 2001;71:S98-S102
© 2001 The Society of Thoracic Surgeons
a Department of Thoracic and Cardiovascular Surgery, Heart Center North Rhine-Westphalia, Ruhr University of Bochum, Bad Oeynhausen, Germany
Address reprint requests to Dr El-Banayosy, Klinik für Thorax- und Kardiovaskularchirurgie, Herzzentrum NRW, Georgstr 11, D-32545 Bad Oeynhausen, Germany
e-mail: abanayosy{at}hdz-nrw.de
Presented at the Fifth International Conference on Circulatory Support Devices for Severe Cardiac Failure, New York, NY, Sept 15–17, 2000.
Abstract
Background. A variety of sophisticated devices have been developed for mechanical circulatory support in patients bridged to cardiac transplantation. Based on 13 years’ experience, we have developed specific protocols for patient selection and management for different devices.
Methods. The principal systems applied in the bridge-to-transplant cohort are the Thoratec ventricular assist device (n = 144, mean duration of support 53 ± 57 days), the Novacor left ventricular assist system (LVAS) (n = 85, mean duration of support 154 ± 15 days), and the HeartMate LVAS (n = 54, mean duration of support 143 ± 142 days). The Thoratec device is used for biventricular assistance or if the duration of support is expected to be less than 6 months. For long-term support, either the Novacor or HeartMate LVAS are preferred.
Results. Despite careful postoperative patient management, this group of patients is prone to a variety of complications. Bleeding occurred in 22% to 35%, right heart failure in 15% to 26%, neurologic disorders in 7% to 28%, infection in 7% to 30%, and liver failure in 11% to 20% of patients. Complications varied with the device applied and the patient’s preoperative condition. A total of 73 patients were discharged from hospital for a mean period of 184 days; this cumulative experience amounted to 37.5 patient-years.
Conclusions. The Novacor and the HeartMate systems offer the additional possibility of discharging patients during support if they fulfill certain criteria. The main reasons for rehospitalization were thromboembolic and infectious complications.
Mechanical circulatory support as a bridge to transplant has become a generally accepted therapeutic option in patients with end-stage heart failure [1–3]. Several sophisticated devices have been developed over the years that differ in terms of location, kind of support, and driving unit; thus they are suitable for different patients with different therapeutic objectives. Consequently, decisions about patient and device management must take into account the unique groups of patients and the varying features of each system. Based on 13 years’ experience, we have developed specific protocols for patient selection and management with respect to different devices and patients. This report briefly describes our device/patient selection criteria and focuses on different management strategies associated with these devices in a bridge-to-transplant setting.
Devices
The devices available at our institution can be classified into four groups by anatomic position. The extracorporeal devices (centrifugal pump, ABIOMED Inc, Danvers, MA) are usually applied for short-term support up to 4 weeks. The paracorporeal systems (Thoratec, Thoratec Laboratories Corp, Berkley, CA, Medos, Medos, Stollberg, Germany) are suitable for midterm support up to about 6 months, with the Medos device being restricted to pediatric patients only. Application of partially implantable systems (Novacor, Baxter, Healthcare Corp, Berkeley, CA, HeartMate, ThermoCardiosystems Inc, Woburn, MA) is indicated for long-term support. The recently available fully implantable LionHeart (Arrow, Reading, PA) is applied for extended support as destination therapy in patients ineligible for heart transplantation.
Based on the expected duration of support and the intention to treat, we developed selection guidelines for these devices for different groups of patients (Table 1). In patients with short-term support devices, hybrid to long-term support systems were recommended when myocardial damage proved irreversible and patients were eligible for heart transplantation.
|
Between September 1987 and August 2000, we performed 524 implantations in 483 patients. The centrifugal pump was implanted in 139 patients, the ABIOMED system in 88 patients, the Thoratec ventricular assist device (VAD) in 144 patients, the Medos VAD in 8 patients, the DeBakey (Micromed, Houston, TX) VAD in 2 patients, the Novacor left ventricular assist system (LVAS) in 85 patients, the HeartMate LVAS in 54 patients, and the LionHeart in 4 patients.
The systems mainly used for the bridge-to-transplant cohort were the Thoratec VAD (n = 144; 21 women, 123 men; aged 11 to 69 years, mean 48.5 ± 14 years; duration of support 3 to 383 days, mean 53 ± 57 days; left ventricular support n = 68, biventricular support n = 65, right ventricular support n = 6, total artificial heart n = 5), the Novacor LVAS (n = 85; 10 women, 75 men; aged 15 to 71 years, mean 51.5 ± 12 years; duration of support 6 to 1,090 days, mean 154 ± 15 days), and the HeartMate LVAS (n = 54; 2 women, 52 men; aged 21 to 69 years, mean 54 ± 11 years; duration of support 1 to 882 days, mean 143 ± 142 days). For statistical purposes, patients supported with Thoratec as a total artificial heart were included in the biventricular group. The implantation technique of the Thoratec VAD as a total artificial heart has been described elsewhere [2].
Device selection in bridge-to-transplant patients
Patients are selected for Thoratec biventricular support if one of the following conditions are present: central venous pressure more than 20 mm Hg and pulmonary arterial pressure (PAP)-central venous pressure gradient less than 4 mm Hg, increased pulmonary vascular resistance (more than 500 dyn · second · cm-5), multiple organ dysfunction, or severe malignant arrhythmias refractory to medical therapy. Multiple organ dysfunction is defined as failure of two organs (kidney, liver, or intestines) in addition to cardiac failure. Other patients receive a Thoratec left ventricular assist device (LVAD) if the duration of support is expected to be less than 6 months. Those patients who are expected to be supported for a longer period and are potential candidates for out-of-hospital support, receive an implantable LVAD (Novacor or HeartMate, provided body surface area exceeds 1.5 m2).
Statistical analysis
Data were analyzed with SPSS for Windows (SPSS Inc, Chicago, IL). Continuous variables are expressed as mean with standard deviation. Binary variables are described as frequency distributions; intergroup differences are analyzed by Fisher’s exact test. Linearized rates of complications are calculated as the number of complications per patient month in a given time frame, with 95% confidence limits.
Postoperative patient management
In the postoperative period, the general guidelines we apply are not different from those applied after other cardiac surgical procedures. Early extubation and mobilization are important, and enteral feeding should be reestablished as soon as possible. Patients are moved to a normal ward early. Nevertheless, patients supported with a VAD constitute a unique cohort who is susceptible to specific complications. As in other cardiac surgical patients, our primary effort is to prevent postoperative problems. From the beginning, careful patient selection and early initiation of mechanical circulatory support are important factors. In addition, meticulous operative technique, stringent operating room and intensive care protocols, and careful strategies for infection prophylaxis and anticoagulation require particular attention. With regard to different devices, protocols for antibiotic prophylaxis and hemodynamic management are basically the same. Antibiotic regimes, however, differ depending on the device (see below on neurologic complications). All of these issues are dealt with by a specialized team that cares for each particular patient group in inpatient and outpatient settings.
A variety of complications may arise, the most important of which are bleeding, right heart failure, neurologic disorders, infection, and liver failure.
Definitions and risk factors for major complications
A bleeding complication is defined as blood loss of more than 1,500 mL/m2 in 24 hours. Right heart failure in patients with left ventricular support is defined as a cardiac index below 2.2 L/min/m2 with a central venous pressure of 18 to 22 mm Hg and double-drug inotropic support in the absence of a high pulmonary vascular resistance for at least 2 hours. Major neurologic complications are defined as a neurologic deficit proved and differentiated by computed tomography scan. A pocket infection is defined as local signs of infection with purulent secretions that require lavage drainage and contain positive bacterial cultures. Criteria for valved conduit endocarditis are signs of systemic infection despite adequate antibiotic therapy, increased central venous pressure, low pump output with a dilated left ventricle, and an abnormal Doppler image above the inflow cannula. The presence of a septic complication was indicated by a body temperature above 38.5°C, white blood count above 12,000 g/dL, high output states, low systemic vascular resistance, and positive blood cultures. Liver failure is diagnosed when a patient’s bilirubin value is more than 8 mg/dL.
In addition to procedure-related issues such as a prolonged duration of cardiopulmonary bypass or massive blood loss and blood transfusion, major patient-related risk factors must be considered before implantation. These factors include poor general condition, liver dysfunction, preoperative right heart insufficiency, prolonged hospitalization, long-term ventilation, immobility, or multiple organ dysfunction. All these factors may contribute to possible development of postoperative complications.
Bleeding
Bleeding represents one of the most frequent complications after implantation of a VAD. Depending on different definitions by authors, the incidence of bleeding is reported to be 20% to 45% [4–7]. Bleeding occurs either as an early complication resulting from the operating technique or as a late complication at connector sites [8]. One of the most important factors in bleeding management is meticulous hemostasis during operation. Aprotinin is administered in patients with bleeding of nonsurgical origin, heparin is replaced by protamine, and fresh frozen plasma and platelets are given when required. According to our anticoagulation protocol, patients do not receive heparin within the first 24 hours postoperatively. Finally, the heart-lung machine is washed before being returned.
The incidence of bleeding in our experience with VADs is listed in Table 2. The differentiation between the period before and after the year 1996 shows our learning curve and reflects a marked reduction of bleeding complications in the last few years. This improvement may have resulted from initiating a wrapping technique for implantation of any nonpreclotted prosthesis and inflow and outflow conduits in the HeartMate system. There was a significantly higher incidence of bleeding among patients with biventricular support (34%, n = 24) than with left ventricular assistance (22%, n = 46) (p = 0.04, Fisher’s exact test). The incidence of other major complications among patients having a bleeding complication is summarized in Table 3. There was a significantly higher incidence of liver failure among patients with bleeding than among those without. When analyzing preoperative hepatic values, gamma-glutamyl transferase and alkaline phosphatase levels were significantly higher in patients with a postoperative bleeding complication than in those without. Glutamic-oxaloacetic acid transaminase, bilirubin, and glutamic pyruvic transaminase levels did not show these differences. Thus, we cannot exclude the possibility that preoperative liver function influences postoperative bleeding.
|
|
In our experience, 15% (n = 10) of patients with Thoratec LVAD, 25% (n = 21) of patients with Novacor LVAD, and 26% (n = 14) of patients with HeartMate LVAD had right heart failure. Right ventricular assistance was required in 7% (n = 6) of Novacor patients (centrifugal pump n = 2, Thoratec RVAD n = 3, Medos n = 1) and 11% (n = 6) of HeartMate patients (centrifugal pump n = 4, Thoratec RVAD n = 1, Medos n = 1). The difference between both systems was not significant. Furthermore, right heart failure occurred more often after urgent or emergent LVAD implantation than after an elective procedure (24% [n = 16] versus 21% [n = 22]; p = 0.5, Fisher’s exact test). However, right heart failure significantly affected outcome, as 35% of patients with right heart failure (n = 15) and 63% of patients without (n = 103) were successfully bridged to transplantation (p = 0.002). In the biventricular assist device (BVAD) group, 39 of 70 (56%) survived to transplantation.
Neurologic complications
Thromboembolism constitutes a major postoperative complication and is always a concern in patients with mechanical circulatory support, because the blood contacting surfaces within the devices is foreign material that increases the risk of thrombus formation. The prevalence of thromboembolism depends markedly on the system used and ranges between 9% and 47% [4, 7, 10]. However, not all neurologic events occurring during support are device-related. A patient group comparable in age and underlying disease (reduced left ventricular function, atrial fibrillation) that does not receive mechanical circulatory support is also susceptible to thromboembolic complications. To prevent neurologic disorders, we follow a careful anticoagulation protocol: In the first 24 hours postoperatively the patients receive no anticoagulants. Thereafter, therapy is started with heparin according to the activated clotting time (1.5 x initial value) in patients with the HeartMate device or according to partial thromboplastin time (1.5 to 2 x initial value) with Thoratec and Novacor. Long-term anticoagulation consists of phenoprocoumon and clopidogrel 75 mg daily with Thoratec and Novacor (international normalized ratio 2.5 to 3.5), and aspirin (100 mg daily) exclusively in HeartMate patients.
Thromboembolic events occurred in 28% (n = 24) of Novacor patients (0.06 events/patient month), 7.4% (n = 4) of HeartMate patients (0.02/patient month), and 15% (n = 22) of Thoratec patients (0.09/patient month) (p = 0.001). Differences between groups were not significant for cerebral bleeding (0.001/patient month with Novacor, 0.004 events/patient month).
To reduce the incidence of thromboembolism in Novacor patients, new gelatin-coated inflow and outflow grafts with a cross-section of 18 mm have been developed recently in our institution, and are expected to provide improved flow patterns. These grafts have been used in 6 patients with promising results; one event of thromboembolism was noted.
Infection
Infection, yet another major complication associated with LVADs, has been reported to have an incidence of 12% to 48% [4, 5, 11, 12]. Patients may have nonsystem-related infections affecting the respiratory or urinary tracts, other organ systems, or intravenous lines that may cause septicemia. System-related infections, however, may develop that affect the exit site, the driveline, or the pocket and lead to a conduit endocarditis.
Prevention of infection includes careful patient selection, operating room discipline, short-term antimicrobial prophylaxis or specific treatment, meticulously sterile dressing changes and local care of exit sites, removal of all intravenous lines and tubes as soon as medically indicated, early extubation and mobilization, and adequate nutrition.
Driveline or exit site infections were found in 24% (n = 20) of Novacor patients (0.05/patient month), in 30% (n = 16) of HeartMate patients (0.06/patient month), and in 9% (n = 13) of Thoratec patients (0.05/patient month); differences between the systems were not significant. A pocket infection occurred in 7% (n = 6) of Novacor patients (0.01/patient month) and in 21% (n = 9) of vented electrical HeartMate patients (0.04/patient month); this difference was significant (p = 0.049). However, because we developed a new method to prevent pocket infections in HeartMate patients in 1998, the incidence of this complication in the HeartMate group declined from 33% to 10% (p = 0.06) [13]. Conduit endocarditis was found in 4% (n = 3) of Novacor (0.007/patient month) and 3.7% (n = 2) of HeartMate patients (0.009/patient month).
Pocket infections usually can be managed by local irrigation. If signs of systemic infection become obvious, antibiotic agents are given for at least 4 weeks. Patients with uncontrolled infection despite the above measures are upgraded to a higher priority for transplantation according to Eurotransplant status class criteria.
Liver failure
Liver failure occurred in 46 of 277 (16.6%) of patients, of whom only 33% (n = 15) were successfully bridged to transplantation, as compared with 60% (n = 137) without liver failure. This difference was significant (p = 0.001). The incidence of liver failure among Thoratec patients was 20% (n = 30), among Novacor patients 12% (n = 10) and in the HeartMate group 11% (n = 6); these differences are not statistically significant. However, there was a significant difference between the incidence of liver failure in LVAD patients (13%, n = 27) and BVAD patients (27%, n = 19) (p = 0.009). In the Thoratec group, liver failure occurred in 27% of BVAD patients (n = 19) and in 15% of LVAD patients (n = 10) (p = 0.09). BVAD patients generally are sicker than those who receive univentricular support (preoperatively higher bilirubin values and worse general situation, more anastomoses during operation, and more bleeding complications). This fact may explain the higher incidence of liver failure.
Management of liver failure normally includes avoidance of drugs with hepatotoxic effects, initiation of thorough drug monitoring to avoid hepatotoxicity, and application of a specific modified diet for the patient.
Eight months ago, we started a randomized study using a new dialysis procedure (MARS = molecular absorbents recirculatory system) to remove liver toxins. Hopefully this procedure will provide therapeutic improvement in the field of liver failure. Otherwise, there is no specific therapy of this complication.
Apart from the complications described above, a variety of other problems occurred in our series and are summarized in Table 4.
|
A major, current issue for the bridge-to-transplant cohort is the possibility of discharging patients home while on the device [14]. This therapeutic option is available mainly for partial assist systems (Novacor, HeartMate).
Before beginning the OOH program in 1994, a team was organized and given responsibility for choosing OOH candidates and teaching these LVAD patients how to be self-sufficient outside the hospital. This training involved LVAD patients and their relatives. The OOH team at our hospital is multidisciplinary and consists of a cardiac surgeon, a cardiologist, an intensivist, and two VAD coordinators. The team provides a 24-hour on-call service that ensures support for patients and families whenever they need it.
Based on 6 years’ experience, we have established the following patient selection criteria for OOH discharge:
Intensive training is started once LVAD patients have moved from the intensive care unit to the ward. Patients and family members are taught how to operate the LVAD system under routine conditions and how to troubleshoot when conditions are irregular. Additionally, patients are taught various diagnostic functions, which include self-testing of international normalized ratio, taking blood pressure and body temperature readings, and caring for the driveline exit site.
To assess partial recovery of the left ventricle, pump output is gradually reduced to 30 bpm. With the pump rate at 30, partial recovery is defined as opening of the aortic valve as assessed by echocardiography when the systolic blood pressure is 80 mm Hg or more.
A total of 73 patients were discharged home with support (Novacor n = 47, HeartMate n = 23, Thoratec TLC-II n = 1, LionHeart n = 2). Their OOH duration of support ranged from 2 to 1,043 days (mean 184 days), with our cumulative experience amounting to 37.5 patient-years. Forty-one patients were rehospitalized mainly for thromboembolic or infectious complications. Readmission rate was 1.8 patient-years. Fifty percent of all patients were free from readmission at 229 days with no significant differences being observed between the Novacor and the HeartMate groups.
Comment
Several factors are important in the management of patients during mechanical circulatory support. First, selection of the appropriate device for the individual patient must take into consideration the intention of treatment, the expected duration of support, and the specific features of the system. The decision whether to initiate biventricular or left ventricular support should be based not only on the hemodynamic situation but also on the status of organ function. Many postoperative problems are probably due to an impaired preoperative condition of the patient. For this reason, we suggest early implantation before multiple organ dysfunction occurs, because liver and right heart failure, in particular, have proved to significantly affect outcome. Postoperatively, prevention of complications is of utmost importance, and in this respect, teamwork of the VAD staff plays a decisive role. Finally, the possibility of discharging patients home while on the device is a worthwhile therapeutic option that should be chosen whenever possible.
Acknowledgments
Grant support was given by the German Association of Organ Recipients (Registered Association).
References
This article has been cited by other articles:
|
|
 |
|
  T. Ota, T. W. Gilbert, S. F. Badylak, D. Schwartzman, and M. A. Zenati Electromechanical characterization of a tissue-engineered myocardial patch derived from extracellular matrix J. Thorac. Cardiovasc. Surg., April 1, 2007; 133(4): 979 - 985. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. Schenk, A. El-Banayosy, W. Prohaska, L. Arusoglu, M. Morshuis, W. Koester-Eiserfunke, L. Kizner, E. Murray, P. Eichler, R. Koerfer, et al. Heparin-induced thrombocytopenia in patients receiving mechanical circulatory support J. Thorac. Cardiovasc. Surg., June 1, 2006; 131(6): 1373 - 1381. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. El-Banayosy, D. Cobaugh, A. Zittermann, L. Kitzner, L. Arusoglu, M. Morshuis, H. Milting, G. Tenderich, and R. Koerfer A Multidisciplinary Network To Save the Lives of Severe, Persistent Cardiogenic Shock Patients Ann. Thorac. Surg., August 1, 2005; 80(2): 543 - 547. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. El-Banayosy, L. Arusoglu, M. Morshuis, L. Kizner, G. Tenderich, P. Sarnowski, H. Milting, and R. Koerfer CardioWest Total Artificial Heart: Bad Oeynhausen Experience Ann. Thorac. Surg., August 1, 2005; 80(2): 548 - 552. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. Haddad, R. G. Masters, P. J. Hendry, A. Kawai, J. P. Veinot, G. Lavallee, and T. V. Mussivand Intercontinental LVAS Patient Transport Ann. Thorac. Surg., November 1, 2004; 78(5): 1818 - 1820. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 F. Collart, H. Feier, D. Metras, and T. G. Mesana A safe, alternative technique for off-pump left ventricular assist device implantation in high-risk reoperative cases Interactive CardioVascular and Thoracic Surgery, June 1, 2004; 3(2): 286 - 288. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 R.D. Dowling, S.J. Park, F.D. Pagani, A.J. Tector, Y. Naka, T.B. Icenogle, V.L. Poirier, and O.H. Frazier HeartMate(R) VE LVAS design enhancements and its impact on device reliability Eur. J. Cardiothorac. Surg., June 1, 2004; 25(6): 958 - 963. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. A. Morgan, R. John, V. Rao, A. D. Weinberg, B. J. Lee, P. A. Mazzeo, M. R. Flannery, J. M. Chen, M. C. Oz, and Y. Naka Bridging to transplant with the HeartMate left ventricular assist device: The Columbia Presbyterian 12-year experience J. Thorac. Cardiovasc. Surg., May 1, 2004; 127(5): 1309 - 1316. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
G. Kleikamp, A. Maleszka, N. Reiss, B. Stuttgen, and R. Korfer Determinants of mid- and long-term results in patients after surgical revascularization for ischemic cardiomyopathy Ann. Thorac. Surg., May 1, 2003; 75(5): 1406 - 1412. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 E. L. Kukuy, M. C. Oz, and Y. Naka Long-Term Mechanical Circulatory Support Card. Surg. Adult, January 1, 2003; 2(2003): 1491 - 1506. [Full Text]
|
|
|
|
|
|
J.M. Grinda, C.H. Latremouille, P. Chevalier, N. D'Attelis, F. Boughenou, R. Guillemain, A. Deloche, and J.N. Fabiani Bridge to transplantation with the DeBakey VAD(R) axial pump: a single center report Eur. J. Cardiothorac. Surg., December 1, 2002; 22(6): 965 - 970. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. Miyagawa, Y. Sawa, S. Taketani, N. Kawaguchi, T. Nakamura, N. Matsuura, and H. Matsuda Myocardial Regeneration Therapy for Heart Failure: Hepatocyte Growth Factor Enhances the Effect of Cellular Cardiomyoplasty Circulation, May 28, 2002; 105(21): 2556 - 2561. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 L. A. Brewer, H. C. M. Lwamba, M. P. Murtaugh, A. C. Palmenberg, C. Brown, and M. K. Njenga Porcine Encephalomyocarditis Virus Persists in Pig Myocardium and Infects Human Myocardial Cells J. Virol., December 1, 2001; 75(23): 11621 - 11629. [Abstract] [Full Text]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| ANN THORAC SURG | ASIAN CARDIOVASC THORAC ANN | EUR J CARDIOTHORAC SURG |
| J THORAC CARDIOVASC SURG | ICVTS | ALL CTSNet JOURNALS |
