Abstract
Objective: The HeartMate II (HM II) LVAD is a small, quiet, continuous-flow, left ventricular assist device (LVAD) for circulatory support in advanced heart-failure patients, with over 2000 implants worldwide. This article reports on the European experience with this device. Methods: The HM II was implanted in 571 patients at 64 European institutions. In 72% of cases (411 patients), implantation has taken place at least 6 months before the closing date of the study (1 August 2008). Patients (19% female, 70% ischaemic aetiology) were on maximum medical therapy, including inotropic support. Body surface area ranged from 1.30 to 2.50 m2 and age from 14 to 75 years (mean: 51 ± 14 years; n = 115, 28% over age 60 years). The intention of support was to provide a bridge to transplantation (73%), destination therapy (21%) and a bridge to recovery (6%). Adverse events were documented in the first 53 patients – for obtaining the Conformité Européenne (CE) Mark (group A) – from a European multicentric study (Strüber et al. [Strüber M, Sander K, Lahpor J, Ahn H, Litzler P-Y, Drakos SG, Musumeci F, Schlensak C, Friedrich I, Gustafsson R, Oertel F, Leprince P. HeartMate II left ventricular assist device; early European experience. Eur J Cardiovasc Surg 2008;34(2):289–94.]: 101 patients) and from a single-centre study (UMCU, The Netherlands: 30 patients). Results: The mean support duration ranged from 0 to 1019 days with a mean of 236 ± 214 days (249 patients: >6 months, 119: 1 year, 12: >2 years; total support time: 293 years). The overall survival to transplantation, recovery or ongoing device support at the end of the study was 69% (284) with an early mortality of 17.5% and late mortality of 13.5%. Of the surviving patients, 23% have been transplanted, 4% had their device removed after recovery of the left ventricle and 42% are still ongoing. Adverse events included bleeding (ranging from 42% in group C to 59% in group A), percutaneous lead infections (A: 0.19, B: 0.61 and C: 0.18 events per patient year), pocket infections (A: 0.08, B: 0.07 and C: 0.09 events per patient year), ischaemic stroke (A: 0.06, B: 0.09 and C: 0.04 events per patient year), haemorrhagic stroke (B: 0.07, C: 0.04 events per patient year) and transient ischaemic attacks (TIAs; A: 0.08, B: 0.02 and C: 0.13 events per patient year). Conclusions: These results support the use of the HM II continuous-flow LVAD for long-term support as a bridge to transplantation and possibly for destination therapy. Future emphasis should focus on minimising adverse events such as infections, bleeding and neurological events.
1 Introduction
Implantable left ventricular assist devices (LVADs) are presently accepted as a successful tool for bridging patients with end-stage heart failure to heart transplantation (BTT). As a result of an increasing shortage of suitable donor organs, waiting lists for heart transplantation (HTx) are growing in most countries as are support durations for VADs in BTT programmes. Although the first generation of pulsatile volume-displacement pumps have shown excellent results as a BTT device, they have their constraints with regard to their large volume and limitations in mechanical durability. The second generation of smaller axial-flow devices, such as the HeartMate II (HM II), are expected to have less of these limitations and, therefore, have an advantage in long-term support over the first-generation devices. This might also lead to more permanent support of the failing left ventricle as an alternative for heart transplantation. Since its introduction in March 2004, there have been over 2000 implants of the HeartMate II worldwide.
We report on the European experience with this device as a bridge to transplantation and as a destination device.
2 Patients and methods
During the period from March 2004 until August 2008, the HM II LVAD has been implanted in 571 patients suffering from end-stage heart failure secondary to cardiomyopathy across 64 European institutions (Databank Thoratec Corp, Pleasanton, CA, USA). In 72% of cases (411 patients), implantation has taken place at least 6 months prior to the closing date of the study (1 August 2008). These 411 patients constitute the study group.
Of the patients studied, 81% were male and 19% were female, with body surface areas ranging from 1.30 to 2.50 m2. The age of patients included in the study ranged from 14 to 75 years with a mean of 51 ± 14 years. Twenty-eight percent of patients were over 60 years of age (n = 115). In the majority of cases, the aetiology of cardiomyopathy was ischaemic (70%). All patients were New York Heart Association (NYHA) class IIIb or IV and were on maximum medical treatment including intravenous inotropic support.
The intention of support was bridge to transplantation (73%), destination therapy (21%) and bridge to recovery (6%).
The device implantation was done as a standard procedure: placement in a pre-peritoneal pocket (majority) or an intra-peritoneal pocket (minority), all according to local clinical practise. The pump was connected to the left ventricular apex and the ascending aorta using extracorporeal circulation (ECC) on the beating heart or during cardioplegic arrest as per the convenience of the surgeon. After weaning from ECC, heparin was antagonised with full-dose protamin. The heparin infusion was re-started after 12–24 h when drain output had reached levels of 30 ml h−1 or less, to achieve partial thromboplastin time (PTT) levels of 50–70 s or an activated clotting time (ACT) of 180 ± 20 s. Anticoagulation with acenocoumarol, with or without aspirin, was carried out on an individual institutional basis.
Adverse events, which had not been documented in the Thoratec Registry, occurred in the first 53 patients in the original HM II Conformité Européenne (CE) Mark study (group A), in 101 patients of a European multicentric study (group B) and in 30 patients in a Dutch single-centre study (group C). Table 1 shows the patient characteristics of groups A, B and C. The outcomes were determined in all 411 patients.
Patient characteristics and LVAD support duration.
Survival to transplantation, recovery of the natural heart or ongoing device support was determined by competing-outcomes analysis. Comparison amongst groups was performed by chi-square analysis for categorical parameters and by analysis of variance (ANOVA) for continuous variables. Adverse event rates were determined as events per patient year of LVAD support.
3 Results
3.1 Outcomes
All 411 patients were followed for a minimum of 180 days or until either transplantation, explantation after recovery or death.
The duration of support ranged from 0 to 1019 days with a mean of 236 ± 214 days and a total of 293 patient-years support time. A total of 249 (61%) were supported for more than 6 months, 119 (29%) for more than 1 year and 12 (3.0%) for more than 2 years. The overall survival to transplantation, recovery of the natural heart function with device removal or ongoing device support was 69% (284) at the end of the study, with an early mortality (≪30 days) of 18% and late mortality of 13%. The survival rate at 6 months was 74% and at 1 year it was 71.5%. Of the surviving patients by the end of the follow-up period, 23% have been transplanted, 4% have had their device removed after recovery of the left ventricle and 42% are still ongoing 6 months following the implant (Fig. 1 ).
Competing-outcomes analysis of survival to transplantation, recovery of the natural heart, or ongoing device support at 1 year.
The actuarial survival of all patients on device at the 1 year end point is shown in Fig. 2 .
Actuarial survival curve at 1 year end point.
In all three groups, A, B and C, the most frequent cause of death was multi-organ failure mainly occurring as a result of septic complications or right-heart failure; the second most common cause was cerebrovascular accidents (CVAs). Technical failure was hardly a reason for fatal outcome (no failure of pump mechanism, only two drive-line disconnections in study group B).
3.2 Adverse events
The most common adverse events occurring following implantation of a HM VAD are shown in Table 2 .
Adverse events and events per patient year ( ).
3.3 Bleeding
Re-thoracotomy or multiple blood transfusions (>6 units per 24 h) due to bleeding, mainly as a result of coagulopathy, is present in all three studies as a major adverse event ranging from the lowest percentage in group C (43%) to the highest in group A (59%). Bleeding, however, did not prove to be a high risk for mortality.
Other frequent adverse events are cardiac arrhythmias, right-heart failure, renal failure and haemolysis. Right-heart failure is defined as a postoperative need for temporary right ventricular mechanical support or inotropic support for more than 14 days following the implantation. As nitric oxide inhalation is frequently used routinely following device implantation, it is not an indicator of right-heart failure. Except for haemolysis, all other events mainly occurred in the first few weeks postoperatively. Haemolysis (two consecutive plasma-free haemoglobin (PlfHb) values >40 mg dl−1) appeared at different times in the course of follow-up with rates of 6–13%. However, in all cases, this phenomenon was transient and lasted for 2–5 days.
3.4 Infections
Isolated percutaneous lead infections were among the most frequently seen in all three studies with incidences of 0.19, 0.61 and 0.18 per patient year. The majority of these infections could be treated successfully locally in combination with systemic antibiotics and were never a cause of mortality.
Although pocket infections were less frequent, they appeared to be a more serious complication with incidences of 0.08, 0.07 and 0.09 events per patient year.
The incidence of sepsis, still one of the most important causes for mortality varied from 0.13% (C) to 0.62 (B)% per patient year.
3.5 Neurological complications
In all three studies, neurological complications occurred primarily in the first 6 weeks following the implantation.
The incidence of ischaemic stroke was 0.06, 0.09 and 0.04 events per patient year in groups A, B and C, respectively, while the incidence of haemorrhagic stroke was 0.05 and 0.04 events per patient year in groups B and C.
The incidences of transient ischaemic attack (TIA) when reported by patients or referring physicians were 0.08, 0.02 and 0.13 per patient year.
4 Discussion
In the past decade, in many European heart-transplantation centres, bridge to transplantation with mechanical circulatory support systems has become an established treatment for patients with end-stage heart failure [2]. In the same time frame, however, the annual number of donor hearts has gradually decreased, resulting in longer support times exceeding 1 year [3]. This has also created the need for an alternative for heart transplantation, such as mechanical circulatory support as destination therapy.
Although the first-generation, pulsatile, volume-displacement pumps such as HM I have shown good results as BTT with acceptable complication rates, their large volume and drive line might be responsible for an increasing mortality due to infection and sepsis, especially after 1 year of support or longer [7]. Moreover, the mechanical durability during long-term support is of great concern, as shown in the landmark Randomized Evaluation of Mechanical Assistance for the Treatment of Congestive Heart Failure (REMATCH) trial [5]. The second-generation, continuous-flow devices like the HM II are smaller and have fewer mobile parts, thereby increasing their mechanical durability [6]. This might be attributed to less infectious complications and fewer mechanical problems during long-term support. In the literature, the reports concerning the results of the second-generation devices are so far limited, being mainly retrospective and the small number of patients studied. In many reports, only a small percentage of patients reach a support duration of more than 1 year.
This study is an attempt to explore the outcomes of a larger group of HM II patients (411 out of 571) in 64 European institutions who were supported for more than 6 months.
Despite the fact that the mean duration of support of 236 ± 214 days (ranging from 0 to 1019 days) is significantly higher, the overall survival to transplantation, recovery or ongoing support (69%) is comparable to the outcomes reported in most studies, while the survival after 6 months is remarkably stable. The higher survival rate in the single-centre study (C: 81%) is not statistically significant and may be attributed more to the preoperative characteristics of patients and less to the institutional experience, as reported by Lietz et al. [4] The incidence of neurological complications (CVA and TIA) is remarkably low, especially beyond the perioperative phase, when compared to the published results of the HM I study [7]. This is comparable with the recent HM II study by Miller et al. [8] and it might indicate that the HM II continuous-flow pump is superior in supporting patients for more than 1 year so far. The absence of any mechanical failure may also support this assertion.
Since bleeding is still a serious problem, with incidences of more than 50%, special care has to be taken with regard to the anticoagulation protocol. This study, like others [1], supports a less aggressive anticoagulation protocol, particularly in view of the near absence of pump thrombosis.
The infectious complications are still a major problem even for the smaller continuous-flow devices. Although the incidence of drive-line infections and sepsis does not differ much from previous studies, the number of pocket infections is low or nearly absent. Drive-line infections – when well treated – were not life threatening, in contrast to sepsis.
In conclusion, the European experience with the HM II continuous-flow pump favours its use for a longer duration of support as BTT and might pave the way for its use as a destination therapy.
5 Study limitations
The Thoratec Registry is retrospective and does not provide adequate information concerning adverse events and causes of mortality. These have to be derived from institutional studies with the HM II, which are limited, however, in terms of the number of patients studied. In view of this limitation, the results of this study should be interpreted with caution. This report demonstrates once more the need for an International European Registry of Mechanical Circulatory Support like the Interagency Registry for Mechanically Assisted Circulatory Support (INTERMACS) sponsored by the National Institutes of Health (NIH) in the United States of America (USA).
Presented at the 22nd Annual Meeting of the European Association for Cardio-thoracic Surgery, Lisbon, Portugal, September 14–17, 2008.
Appendix A
Conference discussion
Dr R. Korfer (Bad Oeynhausen, Germany): Heart failure is, with exceptions, mainly a disease of elderly patients. So I am wondering about the mean age in the patient group ABC because it's below 50 years of age. In our series, it's more than 60. Maybe you can comment on that.
I'm also wondering about the incidence of bleeding problems postoperatively. Do all the different groups you have presented here have the same anticoagulation protocol or what is the definition of bleeding? And was the management more or less dependent on the experience of their groups with a high number of implantation or small group? Because we all know from the REMATCH study, that influenced the results tremendously.
And I was also wondering because 70% of your patients have indication for ischaemic cardiomyopathy. And these young patients, did they have preoperative CABG? Were they redos that could also explain the high incidence of bleeding?
And what was the preoperative condition of these patients? Catecholamines in all patients, that's mandatory, but what about the use of intra-aortic balloon pump or renal failure or ventilatory support?
So you have different results with problems like right-heart failure. How did you manage this?
And finally, a question to quality of life, because that is very important I think nowadays. How many of those patients you presented here are leaving the hospital and are waiting at home for transplantation?
Dr Lahpor: To start with the quality of life, I can only speak, actually, for my own patient population in this regard. And I know that at this moment, more than about 80% of all patients are able to go home and lead a more or less normal life. I can show you pictures of patients who are still at work with their device.
Bleeding problems. Bleeding problems are of course, a matter of definition, and in this particular case, it was meant one had to perform a reoperation because of bleeding or there was a necessity of over six units of blood transfusion in these cases. It is rather high, and I think it's due to the anticoagulation management, and that's still a concern.
I know that there is some agreement that the anticoagulation management in this particular device can be minimised to Coumadin aiming at an INR of 2.0 with 100 mg of aspirin and no heparin used at all after the implantation.
So it might minimise the bleeding problems as well. But still, it is, and you can read it in all studies and all the international studies as well, it is still a matter of concern.
Age. You're right, the age is around 50. It is a little bit low. It is a mixed group, actually. The majority are still bridge transplant patients, especially in my own experience. But also in others, these patients are a little bit younger than the destination group. And maybe that's one of the reasons why the mean age is still a bit low. But I think it will increase in the future, that's for sure.
Dr Korfer: Right-heart failure?
Dr Lahpor: Right-heart failure. The thing is right-heart failure is still a problem, and if you look at it, still about 20, 25% of all patients on an LVAD has a right-heart failure, more or less.
But in my own experience, I don't use an RVAD that much anymore. In the 30 patients I presented here of my own patient population, we never implanted a RVAD in these patients. We were able to manage the right-heart failure mainly by drug therapy.
Although, I must admit, I lost recently one patient, which after rethinking, I should have implanted an RVAD. But in this particular case there were more issues for the fatal outcome.
Dr Korfer: And it is a selective group because you don't mention the use of intra-aortic balloon pump preoperatively or renal failure or …
Dr Lahpor: It is still a mixed group. And, again, I can go deeply into my own experience. Of these 30 patients, there were at least two patients we bridged in a bridge-to-bridge patient. We bridged them with a CE Mark device, and only a few had an intra-aortic balloon pump.
But I must admit, if you have a very good understanding with your referring cardiologist, and I am in the fortunate position that we do have a situation like that, we are a little bit earlier, in an earlier phase aware of the existence of patients probably in need of an LVAD.
So we are involved already in an early phase with potential VAD candidates meaning that maybe that's why at least one in 6-month survival in my own experience is higher because I'm able to implant this device earlier than other people who have patients in a deteriorating state on IABP.
References
Author notes
The author is member of the Thoratec European Advisory Board.
