The European Society of Cardiology
Left ventricular assistance from bridge to transplantation to destination therapy. The Pavia experience
Department of Cardiac Surgery, University of Pavia, Pavia, Italy
* Correspondence: Mauro Rinaldi, MD, Department of Cardiac Surgery, IRCCS Policlinico San Matteo, University of Pavia, 27100 Pavia, Italy. Tel.: +39382503520; fax: +39382503059 (E-mail: m.rinaldi{at}smatteo.pv.it).
Abstract
OBJECTIVE: The increasing number of patients awaiting heart transplant and the shortage of donors led to the development of a variety of left ventricular assist devices (LVAD). We analyse our experience in order to evaluate the efficacy of LVAD as bridge to transplant and the feasibility of permanent implantation.
PATIENTS: The data are drawn from our experience on 50 patients, implanted with the Novacor LVAD and from a limited series of 4 patients implanted with a Lion Heart totally implantable permanent LVAD.
RESULTS: Seventeen patients died on the device, 32 underwent heart transplant (9 died after transplant) and one is still on device. The causes of death were mostly related to cerebrovascular events or multi-organ failure. Cardiac output, wedge pressure, pulmonary vascular resistance and mean pulmonary pressure improved significantly. Cerebrovascular complications occurred mostly during the first 3 months of assistance, whereas the incidence of infections remained constant during the follow-up period. With a mean time of assistance of 211 days, we had only two cases of device malfunction. The four Lion-Heart patients experienced a clear improvement in hemodynamics but at a price of a significant complication rate. Three of them died after 418, 105 and 380 days of assistance. Device malfunction was observed in two cases.
CONCLUSIONS: LVAD Novacor has shown good hemodynamic improvement and reliable mechanical performance and long-term bridging can be considered fairly safe since most complications seem to occur within the first 90 days. Destination therapy with totally implantable devices seems to need further engineering and technical development.
Keywords Heart failure; Ventricular assist device; Mechanical support; Heart transplant; Destination therapy; Bridge to transplant
Introduction
Despite constant improvement of medical therapy for heart failure, the number of patients waiting for heart transplantation is still bigger than donors availability and many patients with a severely compromised hemodynamic status die, waiting for a transplant. Several ventricular assist devices have been developed as bridge to transplantation1–3 to support these patients even for extended period of time. These electrically powered left ventricular assist devices (LVAD) have been developed both as a bridge to transplant, or bridge to recovery in case of temporary reversible myocardial dysfunction.4,5 They are usually fully wearable (the controller and the battery can be hung on the belt) and patients can freely move around, experiencing a good quality of life.6,7 The patients can be initially transferred in a rehabilitation center for a recovery period and then discharged home.8 In our department, we have implanted LVAD Novacor (World Heart Inc., Ottawa, Ont.) in 50 patients with a long mean time of assistance. We have reviewed our results with particular interest for clinical and mechanical complications occurred during the follow-up period.
In recent times, devices had been modified to increase long-term reliability, while advances in technology allowed considerable reduction of pump and controller dimensions. The bridge-to-transplant experience showed the applicability of mechanical circulatory assistance for long periods of time. The limitation of this application were the long-term complication of these devices, most of them being the consequence of percutaneous leads and cables. The positive clinical results of bridge to transplant assistance have consequently led to the concept of permanent support or destination therapy.
Very recently a totally implantable LVAD called LionHeartTM (Arrow International Inc., Reading, PA) became available for clinical use. The peculiarity of this device is that it is designed for complete intra-corporeal implantation. Energy can be transferred across the skin without any percutaneous cable, with an induction coupling system. With our Novacor experience, we could assess the safety of prolonged period of left mechanical assistance and this new technology allowed us for the first time to consider permanent implantation. Patients with definitive contraindication to heart transplantation were selected to be enrolled in this preliminary trial. Institutional Ethical Committee approval was required and a thorough informed consent was obtained from each patient. The present study reports the first Italian clinical experience with the totally implantable left ventricle assist device LionHeartTM in four patients.
Materials and methods
Novacor series
From November 1992, we have implanted the Novacor LVAS in 50 patients (45 males and 5 females) either as bridge-to-transplant (47 pts), destination therapy (2 pts) or bridge-to-recovery (1 pts). The two destination therapy patients had formal contraindication to transplantation (age>65 and colon cancer). In one female patient affected by myocarditis a bridge to recovery was attempted. The mean age of population was 51.6 years (range 29–68 years). Indication for implantation was dilated cardiomyopathy in 31 pts, ischemic cardiomyopathy in 13 pts, post-cardiothomy heart failure in 4 pts, myocarditis and chronic rejection after heart transplantation in the remaining 2 patients. They were all in advanced cardiac failure: 27 patients (54%) were dependent on intravenous inotropes and vasodilators; 13 patients (26%) were on intra-aortic balloon pump and 3 (6%) were on a centrifugal pump. Eleven patients (22%) had undergone previous cardiac surgery (10 coronary artery bypass graft and 1 cardiac transplantation).
Five patients presented with renal failure (creatinine level>2.0 mg/dl) and 10 patients with different degree of liver dysfunction (two of them with associated coagulopathy). Preimplant hemodynamics are shown in Table 1. In 22 patients (44%), a relevant right heart failure (defined as CVP>15 and RV ejection fraction<10%) was present. In the first two cases, the old console system was used. In all the following, the wearable system was applied.
|
Surgical technique
In all cases, we adopted a modified "ortho-dromic" technique, previously described by our group.9 After sternotomy, the pocket was tailored dividing the insertion of the diaphragm to the lower ribs, for an easier bleeding control. On cardiopulmonary bypass (CPB), with or without aortic clamping, apical cannulation was performed first. The device was then easily deaired, with blood flowing in the physiologic direction. The aorta was then tangentially clamped and the outflow conduit anastomosed. Before tying the suture, the final deairing was obtained. This technique allows extreme precision in apical cannulation, easier control of bleeding and accurate deairing of the pump. The only technical problem was an accidental section of a patent graft on the right coronary artery (RCA), which caused right ventricular failure during the weaning from CPB. The re-anastomosis of the graft to the outflow conduit of the pump allowed the recovery of right ventricular function and the weaning from the CPB. In 41 cases (82%), a significant inotropic support was necessary to assist the right ventricular performance and to wean the patient from CPB. In two cases, a right-side centrifugal pump needed to be implanted immediately in OR and in other two cases later on in ICU. Antibiotics protocol was based on vancomycin, norfloxacin and amphothericin. In the first 22 pts, anticoagulation protocol included immediate post-operative heparin infusion, followed by oral warfarin, in order to maintain a INR between 2.5 and 3.5. Aspirin, ticlopidin or dipyridamol were also used. The subsequent eight patients were treated according to the La Pitié protocol,10 and subsequently we returned to the original scheme.
Lion Heart series
From October 2001 to March 2002, four male patients underwent LionHeartTM LVAD implantation for end-stage cardiac failure as destination therapy. The demographic data are illustrated in Table 2. It was an old population with clear contraindication to transplantation. All patients were in NYHA class IV and patient 1 had intravenous inotropes and vasodilators.
|
Surgical technique of implantation is very similar to the one described for the Novacor LVAD. Through the midline incision, the pocket was extended under the right rectus muscle in the right lower quadrant, to accommodate the controller and the battery. An additional subcutaneous small pocket is developed over the right chest wall for the internal coil of the transcutaneous energy transmission system. The electrical conduit of the pump assembly is tunneled to the controller and the remaining connections among the internal coil, the controller and the pump are secured. The external coil is kept in place over the internal coil with a single loop suture. After connecting the inflow and outflow conduit, the compliance chamber is placed in the left pleural space and connected to the pump housing. The access port is placed in a stable subcutaneous pocket along the left chest and its tubing tunneled through the chest wall and secured to the port. At this stage, pumping is initiated keeping the telemetry wand over the controller, and the patient is slowly weaned from CPB.
Results
Novacor series
Cardiac output, pulmonary capillary wedge pressure, mean pulmonary artery pressure and pulmonary vascular resistance statistically improved after assistance. The improvement of the right ventricular ejection fraction was not statistically significant (Table 1).
Hospital mortality after implantation was 13 pts (26%). Cause of death was multi-organ failure in 5 cases, cerebrovascular accident in 5 cases, right ventricular failure in 2 cases and hemorrhagic shock in the last case. Four of 5 patients who died from multi-organ failure were desperately ill: two were on centrifugal pump and one on IABP with high dose inotrope infusion. All four patients who experienced post-operative right ventricular failure requiring right side centrifugal pump died in ICU before transplant. Late mortality accounted for 4 deaths (10.8%) with 2 cerebrovascular accident, 1 sepsis and 1 lung cancer. Actuarial survival rate of patients on device is represented in Fig. 1. Thirty-two pts (65%) underwent heart transplantation and 1 is still on the device. Of the 32 transplanted patients, nine died after the procedure. The survival rate after transplant is 72% and the overall success of bridging to transplantation is 49%. The actuarial survival of patients transplanted after bridging is 65.6±8.3% at 4 years (see Fig. 2) and this figure does not differ from the general heart transplant population survival. Causes of death after transplant were severe surgical bleeding (1 case), multi-organ failure (3 cases), graft failure (3 cases), stroke and anaphylactic shock (1 case each).
|
|
Mean time on device was 211±268 days (median 136 days, range 41–297 days). During this period, 13 patients remained in hospital and 24 (64%) were discharged home or to a rehabilitation centre.
There were 15 cases (30%) of surgical bleeding requiring re-exploration. All type cerebrovascular accidents (including those with negative CT scan and those without permanent clinical lesions) were the main complication. They accounted for 41 early episodes (within 3 months) on 24 pts and 8 late episodes (after 3 months) on 7 pts. These figures gives a linearized rate of 0.34 ep/patientmonths in the early period vs. 0.032 ep/patientmonths in the late period (P<0.05). Actuarial freedom from cerebrovascular accident on assistance is represented in Fig. 3.
|
There were also three peripheral embolisms (due to incorrect anticoagulation management at home) requiring Fogarty procedure.
Major infection was the other relevant problem during assistance with 39 early episodes on 22 pts and 23 late episodes on 12 pts. The linearized infection rate is 0.33 episodes/patient/months in the early period compared to 0.10 episodes/patient/months in the late period. The difference in incidence is not statistically relevant. The actuarial freedom from infection is represented in Fig. 4. Late infection episodes were mainly sustained by infection of the percutaneous drivelines. Eleven patients (22%) experienced at least one infection episode at the site of cable exit. It tended to be a chronic process with sudden exacerbation of pain and fever. Ascending cable infection occurred in 5 pts: in two cases extending to the abdominal pump pocket and requiring surgical debridment. In 3 pts it caused systemic infection (Candida albicans, Staphylococcus aureus, Corynebacterium) with endocarditis of the device valves. In neither case antibiotic treatment was able to resolve the infection and urgent transplantation was necessary to cure 2 pts, while the third one died of multi-organ failure.
|
Two major device malfunctions were observed. In the first case, a natural wear of the pump bearings was observed after 3 years and the patient was transplanted on day 1297. The second case showed a similar problem, occurring earlier during the follow-up, possibly related to uncontrolled hypertension with high residual volumes of the pump. This patient underwent transplant on day 662. Both procedures were carried out electively and patients were discharged. Four patients required cable substitution and three patients required prophylactic controller replacement, due to repeated unclear alarming. All patients were successfully discharged after the procedure.
Lion Heart series
There was not operative mortality. A prolonged inotropic support with dobutamine (mean: 28±17.8 days) was needed in all cases to support right ventricular function. Pulmonary vasodilation was obtained with nipride infusion or inhaled nitric oxide (patient 1). The average length of mechanical ventilation was 12±16 days. Patient 3 underwent percutaneous tracheostomy because of prolonged ventilatory support, whereas patient 4 had to be mechanically ventilated again on post-operative day 7 because of acute respiratory failure. Mean stay in ICU was 16.8±13.6 days. Three patients were discharged home after been physically rehabilitated and trained to take care of the power transmitter and the batteries. To date, mean assistance duration is 280 days (range: 105–450). Hemodynamic data before and after device implantation are summarized in Table 3. The efficacy of the device in normalizing circulatory conditions is clearly evident. Thromboembolic events accounted for 7 episodes in 3 patients (2.3 episodes/pt) with a linearized ratio of 0.13 episodes/patient/months. Five were minor transient ischemic attacks which completely resolved. Two were major strokes, one of which caused patient's death. The second most important post-operative complication was wound dehiscence (3 cases in 4 pts) with consequent infection in the abdominal pocket containing device components. These infections were impossible to eradicate with medical therapy and required surgical debridment. After a short period of healing the dehiscence recurred in two cases with chronic pocket infection. In one case, this infection was the cause of severe sepsis which led patient to death.
|
Two cases of major device malfunction were observed. One controller dysfunction due to recurrent unclear alarms that required controller replacement after 4 months and one pump stop due to rupture of the polyurethane blood sac.
Discussion
Heart failure represents a major public health issue in Western World. Its incidence and prevalence are increasing, as opposed to other cardiovascular diseases. In US, nearly 5 millions of patients are affected by heart failure and 500,000 are diagnosed with it yearly. The disease is responsible of 300,000 patients' death each year. In Europe, 2–10 millions patients are estimated to have heart failure and four-year mortality is 40%. As heart failure is mainly a disease of elderly (approximately 6–10% of people older than 65 years are affected by heart failure) and in Western World mean life span is increasing, the prevalence of this disease is expected to grow in the next future.
Despite recent advances in pharmacological therapy of heart failure, survival and quality of life remain poor. Heart transplantation is known to be an effective therapeutic resource for end stage heart failure. Unfortunately, the scarce number of donors restrains availability of this treatment to a limited number of patients. In US, more than 4250 patients are enrolled in waiting list for heart transplantation, but only 2500 heart transplants are performed yearly. In Italy, in the year 2000, donors fulfilled 41.5% of the need (706 patients on waiting list, 293 heart transplants performed). Thus, the sharp discrepancy between "demand" and "offer" has led to an continuous growth of the number of potential recipients, resulting in longer waiting times and in increased mortality rates (up to 30% of patients die before receiving donor organ).
The shortage of donors and the great number of patients awaiting for heart transplant has led to the increasing use of LVAD. In our experience, the Novacor LVAD was used for those critical cases that could not wait for a donor on medical therapy, and all cases showed a marked improvement of their clinical conditions and hemodynamic parameters.11,12 The right ventricular ejection fraction did not show a statistically significant increase as reported by other authors.13 The quality of life on assistance was reasonable and 24 patients were discharged from hospital.12
Mortality rate on the device was severely influenced by post-operative cerebrovascular events, severe pre-operative conditions, biventricular failure and previous mechanical assistance (IABP and centrifugal pump).14 At least 4 patients who died on the device were in desperate clinical conditions and probably the device was implanted too late. The actuarial survival after implant was 80% at 3 months and 55% at 1 year, satisfactory compared with the natural history of those kind of patients. Quite long was mean time of assistance: 211±268 days, with 12 pts assisted more than 10 months. The explanation of this were the good clinical condition of the assisted patients: at this point in case of a donor availability we preferred to transplant an instable patient.
The actuarial survival after bridging to transplant has been satisfactory (65.6±8.3.% at 4 years) and does not differ from the general transplant population.
Most complications occurred during the first 3 months of assistance and after this period most of the events were infections or a limited number of thromboembolic complications.
We use an aggressive anticoagulation protocol, especially during La Pitié protocol period, and so the incidence of bleeding was relatively low (30%): we were able to manage without problems all the patients but one that died on day 15, after a severe hemorrhagic shock.15
Thromboembolic and hemorrhagic cerebral complications in our series remain the main problems correlated to the LVAD. Our incidence is higher compared with other authors but is similar to other series.14–17 These differing reports can be explained also because we include also all cases with negative CT scan and without permanent clinical lesions. We could not find any correlation between cerebrovascular events and coagulation parameters, different protocols used, different anti-platelet drugs given and different model of device implanted, but the correlation with risk factors for thromboembolism such as infections, atrial fibrillation and atherosclerosis18 remains important.
Contrary to other authors, our statistical analysis shows that in the first 3 months the incidence of CVA was ten time superior than later (0.34 vs. 0.032 episodes/patient/months).19 This situation could be related to an incomplete assessment of anticoagulation or short periods of low cardiac output in the initial time of support.20
The incidence of sepsis and infections was similar to other series.19,20 Percutaneous leads and drivelines are one of the major source of infection which in turn tends to become generalized sepsis. Some late thromboembolic events are strictly correlated with these septic episodes. This tendency renders percutaneous lead a constant hazard for severe systemic infection and valve endocarditis and thromboembolism. As reported by other authors, mediastinitis did not develop in our patients with driveline or abdominal wound infections.19 The incidence rate of infection in our statistical analysis remains constant during the time of assistance. These series of limitations makes this devices not ideal for very prolonged assistance nor for permanent implantation. A totally implantable LVAD should have overcome these severe limitation allowing less complicated prolonged or permanent implantation.
Our preliminary experience with the Lion Heart did not show that these problems are solved: the difficult healing of the surgical wound could be related to the abundant mass of foreign material that needs to be implanted. Three out of 4 patients experienced this complication: even after surgical debridment, sterilization and second closure of the wound the problem presented again as chronic dehiscence and re-infection. In this case, heat production from the pump, the controller and the battery may have prevented the wound from definitive healing. Despite the efficacy of the assistance, the discomfort related to the bulky system cannot be overlooked. A further miniaturization of the controller and the battery is expected in the future, in the attempt to improve these aspect. This new device is not free from thromboembolism with a rate to be determined.
References
- De Vries WC, Anderson JL, Joice LD, et al. Clinical use of the total artificial heart N Engl J Med 1984;13:124-129.
- Ferrar DJ. Thoratec ventricular assist device principal investigator: Preoperative predictors of survival in patients with Thoratec Ventricular Assist Device as a bridge to heart transplantation J Heart Lung Transplant 1994;13:93-97.[Medline]
- Arabia FA, Copeland JG, Smith RG, et al. International experience with the Cardiowest total artificial heart as a bridge to heart transplantation Eur J Cardiothoracic Surg 1997(Suppl 11):5-10.
- Portner PM, Oyer PE, Pennington DG, et al. Implantable electrical left ventricular assist system: bridge to transplantation and the future Ann Thorac Surg 1989;47:142-150.[Abstract]
- Burton NA, Lefrak EA, Macmanus Q, et al. A reliable bridge to cardiac transplantation: the TCI left ventricular assist device Ann Thorac Surg 1993;55:1425-1431.[Abstract]
- Winowich S, Nastala CJ, Pristas JM, et al. Discharging patients who are undergoing mechanical circulatory support Ann Thorac Surg 1996;61:478-479.
[Abstract/Free Full Text] - McCarthy PM, Savage RM, Fraser CD, et al. Hemodynamic and physiologic changes during support with an implantable left ventricular assist device J Thorac Cardiovasc Surg 1995;109:409-418.
[Abstract/Free Full Text] - Viganò M, Scuri S, Corbelli F, et al. Staged discharge out of hospital of the Novacor left ventricular assist system (LVAS) recipients Eur J Cardiothoracic Surg 1997(suppl 11):45-50.
- Viganò M, Martinelli L, Minzioni G, et al. Modified method for Novacor left ventricular assist device implantation Ann Thorac Surg 1996;61:247-249.
[Abstract/Free Full Text] - Copeland J, Szefner J. Anticoagulants and the artificial heartIn: Lewis T, Graham TR, editors. Mechanical circulatory support. London: Edward Arnold; 1995. pp. 306-311.
- McCarthy PM, Portner PM, Tobler HG, et al. Clinical experience with the Novacor ventricular assist system J Thorac Cardiovasc Surg 1991;102:578-587.[Abstract]
- Kormos RL, Murali S, Dew MA, et al. Chronic mechanical circulatory support: rehabilitation, low morbidity, and superior survival Ann Thorac Surg 1994;57:51-58.[Abstract]
- Gallagher RC, Kormos RL, Gasior T, et al. Univentricular support results in reduction of pulmonary resistance and improved right ventricular function ASAIO Trans 1991;37:287-288.
- Schmid C, Weyand M, Nabavi DG, et al. Cerebral and systemic embolization during left ventricular support with the Novacor N100 device Ann Thorac Surg 1998;65:1703-1710.
[Abstract/Free Full Text] - Vetter HO, Kaulbach HG, Schmitz C, et al. Experience with the Novacor left ventricular assist system as a bridge to cardiac transplantation, including the new wearable system J Thorac Cardiovasc Surg 1995;109:74-80.
[Abstract/Free Full Text] - Nabavi DG, Georgiadis D, Mumme T, et al. Clinical relevance of intracranial microembolic signals in patients with left ventricular assist devices Stroke 1996;27:891-896.
[Abstract/Free Full Text] - Houel R, Moczar M, Clerin V, et al. Pseudointima in inflow conduits of left ventricular assist devices Ann Thorac Surg 1999;68:717-723.
- Didisheim P, Olsen DB, Farrar DJ, et al. Infection and thromboembolism with implantable cardiovascular devices ASAIO Trans 1989;35:54-70.[Medline]
- Pennington DG. Extended support with permanent system: percutaneous versus totally implantable Ann Thorac Surg 1996;61:403-406.
[Abstract/Free Full Text] - Kormos RL, Borovetz HL, Pristas JM, et al. LVAS pump performance following initiation of left ventricular assistance ASAIO Trans 1990;36:703-705.
CiteULike 
Connotea 
Del.icio.us What's this?
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||