Download - Change Location

EJCTS-6818; No of Pages 9
European Journal of Cardio-thoracic Surgery xxx (2009) xxx—xxx
www.elsevier.com/locate/ejcts
European experience of DuraHeartTM magnetically levitated
centrifugal left ventricular assist system§,§§
Michiel Morshuis a, Aly El-Banayosy a, Latif Arusoglu a, Reiner Koerfer a,
Roland Hetzer b, Georg Wieselthaler c, Alain Pavie d, Chisato Nojiri e,1,*
a
Heart & Diabetes Center, North Rhine-Westphalia, Bad Oeynhausen, Germany
b
German Heart Institute, Berlin, Germany
c
University of Vienna, Vienna, Austria
d
Pitie Salpetriere Hospital, Paris, France
e
Terumo Heart, Inc., Ann Arbor, MI, USA
Received 1 September 2008; received in revised form 14 December 2008; accepted 16 December 2008
Abstract
Objective: The DuraHeart (Terumo Heart, Inc., Ann Arbor, Michigan, USA) is the world’s first approved magnetically levitated centrifugal left
ventricular assist system designed for long-term circulatory support. We report the clinical outcomes of 68 patients implanted with the DuraHeart
as a bridge to cardiac transplantation in Europe. Methods: Sixty-eight patients with advanced heart failure (six females), who were eligible for
cardiac transplantation were implanted with the DuraHeart between January 2004 and July 2008. Median age was 58 (range: 29—74) years with
31% over 65 years. Thirty-three of these patients received the device as a part of the European multi-center clinical trial. Survival analyses were
conducted for 68 patients and other safety and performance data were analyzed based on 33 trial patients. Results: Mean support duration was
242 243 days (range: 19—1148, median: 161) with a cumulative duration of 45 years. Thirty-five patients (51%) remain ongoing, 18 transplanted,
1 explanted, and 14 died during support with a median time to death of 62 days. The Kaplan—Meier survival rate during support was 81% at 6
months and 77% at 1 year. Of the 13 patients (21%) supported for >1 year, 4 supported for >2 years, 1 supported >3 years, 2 transplanted, 2 died,
and 9 ongoing with a mean duration of 744 216 days (range: 537—1148, median: 651). Major adverse events included driveline/pocket
infection, stroke, bleeding, and right heart failure. There was no incidence of pump mechanical failure, pump thrombosis, or hemolysis.
Conclusions: The DuraHeart was able to provide safe and reliable long-term circulatory support with an improved survival and an acceptable
adverse event rate in advanced heart failure patients who were eligible for transplantation.
# 2008 European Association for Cardio-Thoracic Surgery. Published by Elsevier B.V. All rights reserved.
Keywords: Left ventricular assist device; Bridge to cardiac transplantation; Rotary blood pump; Magnetic levitation
1. Introduction
The first generation of implantable left ventricular assist
device (LVAD) using a pulsatile volume-displacement pump
technology has become an established therapeutic option for
advanced heart failure patients as a bridge to cardiac transplantation. These devices greatly contributed to the advancement of the mechanical circulatory support by providing
improved survival and better quality of life; however there
have been technological limitations, which include a high
thromboembolic rate, mechanical failures associated with
§
Presented at 22nd Annual Meeting of the European Association for CardioSurgery, Lisbon, Portugal, September 14—17, 2008.
§§
The European clinical trial of the DuraHeart LVAS was sponsored by Terumo
Heart, Inc.
* Corresponding author. Tel.: +1 734 7416114; fax: +1 734 7416270.
E-mail address: [email protected] (C. Nojiri).
1
Employee of Terumo Heart, Inc.
moving parts, large size of the implantable pump, and a high
rate of infections [1—3].
Smaller rotary blood pumps emerged as possible alternatives to the large pulsatile devices and these new pumps
eliminated the need for the external venting and prosthetic
valves required for implantable pulsatile pumps. The rotary
blood pumps are categorized into two generations, 2nd and 3rd
generation LVADs based on their technological aspects. Second
generation devices are based on an axial flow pump technology
with a blood-immersed bearing or a pivotal bearing [4—6].
The technical advancement that defines the 3rd generation
LVAD is the elimination of all mechanical contacts inside the
blood chamber. This was accomplished by employing either a
magnetic bearing (magnetic levitation) or a hydrodynamic
bearing between the impeller and the housing. The potential
advantages of a frictionless impeller rotation using these
technologies are reduced hemolysis, reduced potential pump
thrombosis, and an increased mechanical durability necessary
for long-term support. Several manufacturers are in various
1010-7940/$ — see front matter # 2008 European Association for Cardio-Thoracic Surgery. Published by Elsevier B.V. All rights reserved.
doi:10.1016/j.ejcts.2008.12.033
DOCTOPIC: 27
Please cite this article in press as: Morshuis M, et al. European experience of DuraHeartTM magnetically levitated centrifugal left
ventricular assist system. Eur J Cardiothorac Surg (2009), doi:10.1016/j.ejcts.2008.12.033
EJCTS-6818; No of Pages 9
M. Morshuis et al. / European Journal of Cardio-thoracic Surgery xxx (2009) xxx—xxx
2
stages of development and clinical application of 3rd generation rotary blood pumps [7,8].
The DuraHeartTM left ventricular assist system (LVAS)
(Terumo Heart, Inc., Ann Arbor, Michigan, USA) is a 3rd
generation implantable LVAD which combines the magnetic
levitation technology and a centrifugal pump [9—12]. We
report the clinical outcome of the European clinical experience of the DuraHeart LVAS, which include 33 patients from
European multicenter clinical trial and 35 patients after
market launch in Europe.
2. Device description
The DuraHeart LVAS is comprised of three principal
implantable components, a blood pump with a driveline
that connects the pump to an external system, inflow and
outflow conduits, and an externally worn controller and
two battery packs (Fig. 1A). The DuraHeart pump is an
implantable centrifugal pump made of titanium and stainless
steel. The pump uses magnetic levitation to suspend its
rotating impeller within the blood chamber and is composed
of four parts, a magnetic bearing, an impeller, a housing, and
a DC brushless motor (Fig. 1B). The impeller is rotated by a
magnetic coupling between the impeller and the motor, and
is suspended magnetically by the three electromagnets. The
electric current of each electromagnet is controlled by using
three position sensors to keep the impeller free-floating at
the center of the pump housing. The pump requires neither a
rotating shaft nor shaft seals. The pump is housed in a
titanium enclosure that hermetically seals the electrical
components against any blood or tissue contact. The
DuraHeart pump has a diameter of 72 mm, thickness of
45 mm, and weighs approximately 540 g. The displacement
volume is 196 ml, which is 30—50% smaller than the 1st
generation pulsatile pumps. The horizontal length including
the pump and inflow and outflow conduit is approximately
50—60% shorter than that of axial flow LVADs. The DuraHeart
pump is capable of providing 8 l/min of blood flow at
120 mmHg of head pressure with no residual left ventricular
function and is able to generate a wide pressure range
from 50 mmHg at 1200 rpm to 180 mmHg at 2400 rpm, a
comparable capacity to pulsatile pumps. The blood contacting surfaces of the pump, impeller and the inflow and outflow
conduit connectors were entirely modified with a covalently
bound heparin to enhance blood compatibility and reduce
the risk of thrombus formation in low flow areas. An added
feature of the DuraHeart LVAS is the integrated flowestimation algorithm based on the very stable motor current
required to maintain the set rotational speed and viscosity of
the blood. The DuraHeart LVAS was also designed with
additional safety features such as a hydrodynamic bearing
support for back-up in case of failure of magnetic levitation.
The stable impeller position at the center of the pump
housing controlled by the magnetic bearing has an additional
advantage of maintaining consistent washout within the
blood path at all times, thereby reducing the potential of
thrombus formation (Fig. 1B).
The controller is used to establish, maintain and monitor
proper operation of the pump by controlling power inputs to
both the motor and levitation circuits, and by monitoring
output signals for error conditions. The controller can be
powered by wearable batteries, the hospital console, or the
battery charger. The controller also monitors many parameters for fault detection and data logging with 30 days data
storage capacity.
The wearable battery pack is composed of rechargeable
lithium ion battery cells, and two redundant battery packs
provide the power to the controller and the pump for 7 h at
nominal condition. Electronics in the battery provide safety
monitoring and charge status monitoring.
3. Methods
3.1. Study design
Fig. 1. (A) The pump with the inlet and outflow conduits (left) and anatomical
configuration of the implantable and wearable components of the DuraHeartTM
LVAS (right). (B) Schematic illustration of pump components (left) and diagram
of the mechanism of the DuraHeart magnetically levitated centrifugal pump
(right).
3.1.1. CE-mark study
The study was conducted in four centers in Germany,
Austria, and France between January 15, 2004 and March 7,
2007, and was a prospective, multicenter, nonrandomized
trial designed to characterize the safety and performance of
the DuraHeart LVAS in patients eligible for cardiac transplantation. The primary endpoint was to evaluate survival of
patients either to cardiac transplantation or at 13 weeks (3
months) of device support. Secondary endpoints evaluated
adverse events, device performance, and overall patient
status throughout the period of DuraHeart support. The
Please cite this article in press as: Morshuis M, et al. European experience of DuraHeartTM magnetically levitated centrifugal left
ventricular assist system. Eur J Cardiothorac Surg (2009), doi:10.1016/j.ejcts.2008.12.033
EJCTS-6818; No of Pages 9
M. Morshuis et al. / European Journal of Cardio-thoracic Surgery xxx (2009) xxx—xxx
patients who met discharge criteria were allowed to be
discharged from the hospital.
Prior to the start of the clinical investigation, the study
protocol and patient informed consent documents were
reviewed and approved by the ethics committee at each
investigational site. Written informed consent was obtained
from all patients. A clinical events committee (CEC) comprised
of study investigators and independent physicians reviewed
and adjudicated all adverse events and deaths.
The study was conducted in compliance with the following
guidelines and standards: Declaration of Helsinki (Tokyo,
October 2004), ISO 14155: Parts 1 and 2 (2003), ICH Guideline
for Good Clinical Practice.
3.1.2. Post-market study
After CE-mark approval, the post-market study was
initiated in accordance with MEDDEV 2.12.2 (2004) that
includes 50 subjects from August 2007.
3.2. Patients
Eligible patients for the CE-mark study were adults with
end-stage left ventricular failure, at imminent risk of dying.
Patients were receiving maximal conventional therapy, but
remained unresponsive to this therapy. All patients had to be
eligible for cardiac transplantation and fulfilled the inclusion
and exclusion criteria. The inclusion criteria included the New
York Heart Association (NYHA) functional class IV; a body
surface area of at least 1.1 m2; a cardiac index (CI) of 2.2 L/
min/m2 or less with either a systolic blood pressure of
80 mmHg or less, or a pulmonary capillary pressure (PCWP) of
18 mmHg or greater; patients were receiving optimal medical
treatment, including inotropes and/or intra-aortic balloon
pump (IABP). Table 1 summarizes inclusion and exclusion
criteria.
The eligibility criteria for the post-market study were
similar to those for the CE-mark study except that there were
no hemodynamic criteria in the post-market study.
3.3. Implant and explant procedures
Similar to other LVADs, the DuraHeart LVAS was implanted through a median sternotomy with the use of cardiopulmonary bypass (CPB). A small abdominal pocket
between the posterior rectus sheath and fascia transversalis was created for pump placement. After initiation of
CPB, the left ventricular apex was cored and the apical
cuff was sutured onto the left ventricular apex using
pledgetted mattress sutures placed circumferentially. The
inflow conduit was inserted through the apical cuff and
firmly secured by using heavy ligatures, then connected to
the pump inlet port. The outflow graft was connected to
the pump outlet port and anastomosed end-to-side to
the ascending aorta. Deairing from the pump was performed through an 18-gauge needle placed in the outflow
graft. The pump cable was tunneled subcutaneously to
exit the skin at the right lower quadrant and connected
to the controller. The pump rotation was stated to obtain
an optimal flow rate while CBP was gradually discontinued.
Upon transplantation, the following explant procedure
was performed. The original incision was re-opened and
3
Table 1
Patient selection criteria.
Inclusion and exclusion criteria
Inclusion criteria
Patient referred for, and eligible for, cardiac transplantation
Body surface area (BSA) 1.1 m 2
NYHA Functional Class IV
Cardiac index (CI) 2.2 l/min/m2 with either
Systolic blood pressure 80 mmHg or
LAP (PCWP) or PAD 18 mmHg.
Receiving optimal medical treatment, including inotropes and/or IABP
Gives informed consent
All laboratory and physiologic data used for evaluation of patient
status were collected within 48 h of enrollment
Exclusion criteria
Surgical contraindications to LVAD implantation
High risk cardiothoracic surgery within 30 days of enrollment
Myocardial infarction within 30 days of enrollment
Aortic regurgitation grade 1
Evidence of recent or life-limiting malignant disease
Patients with either an implanted mechanical aortic or mitral heart valve
Fixed pulmonary hypertension with a PVR 480 dynes-sec/cm 5
Severe COPD as evidenced by FEV1 1.5 l/min
On ventilator support for 1 week within 30 days of enrollment
Presence of active systemic infection
Symptomatic cerebrovascular disease
Serum creatinine > 5 mg/dl or BUN > 100 mg/dl or requires hemodialysis
Liver enzyme (SGOT, SGPT, or LDH) > 3 times upper limit of normal or
total bilirubin > 5 mg/dl, or biopsy-proven liver cirrhosis or portal
hypertension
Body surface area (BSA) < 1.1 m 2
Diagnosed primary coagulation or platelet dysfunction
Primary RV dysfunction
High probability of noncompliance with protocol
Previous circulatory support, except for IABP
Irreversible end-organ dysfunction, except for the heart
Patient of minor age (as defined by national law)
Pregnancy
Currently participating in another investigational clinical study
Does not understand the aim of the study or does not agree to cooperate
CPB was initiated. The pump was stopped using the
hospital console connected to the controller. The outflow
graft was cross-clamped distal to the aortic anastomosis
and excised. After cutting the ligatures around the apical
cuff, the inflow conduit was removed from the left ventricle. The pump cable was cut within the pocket and the
distal end of the cable was pulled out through the skin exit
site. The pump with both inflow and outflow conduits
and the cable were put in the container, and then shipped
back to the manufacturer for engineering and histological
analyses.
3.4. Anticoagulation
The anticoagulation regimen recommended in the protocol includes intravenous heparin perioperatively followed by
oral coumadin and aspirin. Heparin is administered 8—12 h
postoperatively with a prothrombin time (PTT) of 50—80 s
unless significant bleeding occurs. Once the patient is able to
take oral medication, coumadin is administered to maintain
an international normalized ratio (INR) of 2.5—3.5, along
with 100 mg of aspirin daily. One of the investigational sites
implemented less intensive anticoagulation with coumadin
to maintain INR of 2.0—2.5.
Please cite this article in press as: Morshuis M, et al. European experience of DuraHeartTM magnetically levitated centrifugal left
ventricular assist system. Eur J Cardiothorac Surg (2009), doi:10.1016/j.ejcts.2008.12.033
EJCTS-6818; No of Pages 9
M. Morshuis et al. / European Journal of Cardio-thoracic Surgery xxx (2009) xxx—xxx
4
3.5. Adverse event definition
The adverse event definitions were based upon the
Interagency Registry for Mechanical Assisted Circulatory
Support (INTERMACS, http://www.intermacs.org). All adverse
events were adjudicated by the CEC for device relatedness.
the number and percent of patients with the event, the
number of events, and event rates per patient-year.
For the post-market study, only survival and partial
baseline patient characteristics analyses were conducted
due to a limited availability of the data that had been
monitored and adjudicated.
3.6. Statistical analysis
4. Results
The database for the CE-mark study was closed on June
15, 2007 and the data were monitored and verified for
statistical analyses. All 33 CE-mark study patients were
followed for at least 3 months or until either transplant or
death at the time of database closure and the follow-up
continued until now. Analyses were conducted on the
intention-to-treat population using SAS statistical software
program version 8.2 for Windows (SAS Institute, Cary, NC) by
independent statisticians. The Kaplan—Meier method was
used to estimate the survival rate over time with censoring
for transplant or device explant. A log rank statistic was used
for comparison. Continuous variables were presented as
mean standard deviation (SD) or median, and range. For
comparison, the Mann—Whitney test was used. The level of
statistical significance was set at p < 0.05.
Adverse events were analyzed only for CE-mark study
patients at the time of database closure on June 15, 2007,
which have been adjudicated by the CEC. The data included
4.1. Patients
A total of 68 patients were implanted with the DuraHeart
between January 2004 and July 2008. Of those, 33 patients
who met the inclusion and exclusion criteria were enrolled in
CE-mark study and 35 patients were enrolled in post-market
study in 4 centers. Forty-eight patients were enrolled at
Heart & Diabetes Center, North Rhine-Westphalia, 14 were at
German Heart Institute Berlin, 5 were at University of
Vienna, and 1 was at Pitie Salpetriere Hospital. The baseline
patient demographics and characteristics of both groups are
shown in Table 2. All patients were rated as New York Heart
Association (NYHA) functional class IV. The mean age was
57 11 years (range: 29—74, median: 58) with a mean body
surface area (BSA) of 1.9 0.2 m2 (range 1.4—2.4 m2,
median: 1.9). In the overall groups, 29 patients (43%) were
older than 60 years and 21 patients (31%) were older than 65
Table 2
Baseline characteristics of the trial and post-market patients.
Characteristics
All (n = 68)
Trial (n = 33)
Post-market (n = 35)
Age year (median)
Male (%)
Body surface area (m2)
NYHA class IV (%)
Ischemic cause of heart failure (%)
Left ventricular ejection fraction (%)
Left ventricular end-diastolic dimension mm
56.7 11.3 (57.6)
90
1.9 0.2
100
51
20.2 6.9
74.6 11.6
55.5 12.5 (57.0)
85
1.9 0.2
100
42
20.2 6.7
74.7 12.9
57.8 10.4 (58.2)
96
2.0 0.2
100
62
20.2 7.0
74.3 9.8
Arterial blood pressure (mmHg)
Systolic
Diastolic
Cardiac index (l/min/m2)
Pulmonary capillary wedge pressure (mmHg)
Central venous pressure (mmHg)
Pulmonary vascular resistance (dyne-sec/cm 5)
Blood chemistry values
Serum albumin (g/dl)
Serum sodium (mmol/l)
Serum creatinine (mg/dl)
Blood urea nitrogen (mg/dl)
Serum alanin aminotransferase (U/l)
Serum aspartate aminotransferase (U/l)
Serum lactate dehydrogenase (mg/dl)
Serum total bilirubin (mg/dl)
Hematologiv values
Hematocrit (%)
Platelets (per mm3)
International normalized ratio
98.1 16.6
61.0 13.2
—
—
—
—
3.4 1.0
132.9 10.4
1.5 0.7
39.1 24.7
80.9 231.1
96.7 347.3
313.7 163.1
1.5 1.6
36.2 6.3
198,000 82,000
1.4 0.5
97.2 16.2
59.2 16.2
1.8 0.31
22 6.7
10.0 4.8
265 98
3.5 1.0
131.2 12.3
1.5 0.5
38.7 26.0
88.9 285.6
114.9 429.3
294.6 120.3
1.4 1.1
37.1 6.7
202,000 91,000
1.5 0.6
99.3 21.7
63.2 15.6
—
—
—
—
3.0 0.2
135.4 6.0
1.6 1.0
39.7 23.4
67.5 91.8
62.4 53.4
344.9 215.9
1.7 2.2
34.8 5.5
193,000 69,000
1.3 0.3
Intravenous inotropic support (%)
—
97
—
Mechanical support prior to implant surgery (%)
ICD/biventricular pacemaker
Intraaortic balloon pump
Mechanical ventilation
—
—
—
82
18
6
—
—
—
Please cite this article in press as: Morshuis M, et al. European experience of DuraHeartTM magnetically levitated centrifugal left
ventricular assist system. Eur J Cardiothorac Surg (2009), doi:10.1016/j.ejcts.2008.12.033
EJCTS-6818; No of Pages 9
M. Morshuis et al. / European Journal of Cardio-thoracic Surgery xxx (2009) xxx—xxx
years. Six (10%) were female and 51% of the patients had
ischemic etiology for heart failure. There was no significant
difference in baseline characteristics between the two
groups.
The use of mechanical devices, cardiac surgery or cardiac
intervention prior to the implantation were analyzed for 33
CE-mark trial patients whose data had been monitored and
verified. Thirty-two patients (97%) received intravenous
inotropic support and 6 (18%) were supported by intra-aortic
balloon pumping (IABP), and 2 (6%) were on mechanical
ventilation. Twenty-seven patients (82%) had an implantable
cardioverter-defibrillator (ICD) and/or biventricular pacemaker. Six patients (18%) had undergone previous cardiac
surgery and 5 (15%) had a history of percutaneous coronary
intervention.
4.2. Survival outcomes
All 68 patients of both CE-mark and post-market studies
were followed until August 25, 2008. Of the 33 CE-mark
trial patients, 27 (82%) reached the primary endpoint of
transplantation or survival at 13 weeks endpoint. Two
patients who were replaced with a second DuraHeart device
and kept alive were included as survivors in the survival
analysis. The Kaplan—Meier survival for the CE-mark study at
the time of database closure of June 15, 2007 was 81% (95%
confidence interval (CI): 63—91) at 13 weeks endpoint and
76% (95% CI: 55—88) at I year. The follow up data as of August
25, 2008 for the CE mark trial were as follows. Three
additional deaths occurred at 239, 549 and 550 days. The
mean support duration was 338 311 days (range: 17—1148,
median: 201). Fourteen patients were transplanted with a
mean time to transplantation of 185 148 days (range: 43—
497, median: 142). Nine patients (27%) still remained on
device support with a mean duration of 744 216 days
(range: 537—1148, median: 651) with 13 patients supported
>1 year. Of these 13 patients, 4 were supported >2 years and
1 was supported >3 years. Kaplan—Meier survival estimates
5
for the CE-mark study were 81% (95% CI: 63—91) at 3 months,
77% (95% CI: 58—89) at 6 months, 72% (95% CI: 51—85) at 1
year, and 57% (95% CI: 31—76) at 2 years (Fig. 2).
Of 35 post-market study patients, 5 (14%) underwent
transplantation, 26 (74%) remain on device with a mean
duration of 170 90 days (range: 19—355, median: 146), 4
(11%) died, and 1 (3%) explanted and exchanged to a total
artificial heat. One patient who underwent device explant
and received a total artificial heart and kept alive was
censored at the time of device explant. Kaplan—Meier
survival for the post-market patients was 94% (95% CI: 78—98)
at 3 months, and 84% (95% CI: 62—94) at 6 months (Fig. 2).
The overall Kaplan—Meier survival estimate of all 68
patients was 87% (95% CI: 77—94) at 3 months, 81% (95% CI:
67—89) at 6 months, 77% at 1 year, and 61% (95% CI: 34—78) at
2 years (Fig. 2).
The cause of death and device relatedness was
adjudicated by the CEC for the CE-mark trial patients as
of June 15, 2007. Table 3 summarizes the cause of death and
comorbidities at the time of death for the CE-mark study.
Seven deaths occurred during device support, six before the
primary endpoint and one after the primary endpoint. The
median time to death was 29 days. The majority of patients
who died had multiple comorbidities. One ischemic and 3
hemorrhagic cerebrovascular accidents (CVA) were determined to be the cause of death in 4 patients (57%) and the
majority of deaths (6 patients, 85%) occurred in the initial
11 patients enrolled in the study. The pump and inflow/
outflow conduits from the patients who died of CVA
were thoroughly analyzed for thrombus formation, signs
of infection, or any abnormality. There was no evidence of
pump thrombosis or infection in the blood path. Due to a
high incidence of fatal intra-cerebral bleeding and other
bleeding complications observed in the initial 11 patients,
the anticoagulation and anti-platelet regimen was reviewed
The investigators then agreed to follow less intensive
anticoagulation/anti-platelet regimen recommended in the
protocol for the remainder of the study.
Fig. 2. Kaplan—Meier analysis of survival after implantation of the DuraHeartTM LVAS for CE-mark trial (n = 33), post-market study (n = 35) and all 68 patients. The
patients were censored when they underwent transplantation or device explant.
Please cite this article in press as: Morshuis M, et al. European experience of DuraHeartTM magnetically levitated centrifugal left
ventricular assist system. Eur J Cardiothorac Surg (2009), doi:10.1016/j.ejcts.2008.12.033
EJCTS-6818; No of Pages 9
M. Morshuis et al. / European Journal of Cardio-thoracic Surgery xxx (2009) xxx—xxx
6
Table 3
Summary of deaths during support.
Implant order
Age
Time to death (day)
Cause of death
Other complications
Device relatedness
#2
67
28
Ischemic CVA
Possibly related
#5
#6
60
63
29
21
#7
73
17
Carotid stenosis, confusion,
Possibly related
#10
#11
#16
66
56
61
86
37
178
Hemorrhagic CVA a
Subdural hematomaa
(non-traumatic bleeding)
Cardiovascular failure
(traumatic fall)
Hemorrhagic CVA a
Hemorrhagic CVA a
Sepsis
HIT II, Left atrial thrombus,
chronic atrial fibrillation
Multi-organ failure
Multi-organ failure
Sepsis (Staphylococcus aureus)
Sepsis (Candida albicans)
Multi-organ failure
Possibly related
Possibly related
Unrelated
Possibly related
Unrelated
a
Three hemorrhagic CVAs with massive intra-cerebral bleeding and one subdural hematoma resulted in immediate deaths; likely associated with excessive
anticoagulation therapy.
leveled off near the upper normal limit for the duration of
the study period. Average plasma-free hemoglobin values
remained within normal ranges throughout the study period.
These data suggest that clinically significant hemolysis did
not occur as a result of implantation of the DuraHeart.
All of the patients enrolled in the CE-mark trial had NYHA
class IV at baseline. At the study endpoint of 13 weeks, 27
patients were supported by the DuraHeart. Six patients (22%)
without assessment of the NYHA class were discharged home
by the 13 weeks endpoint. Of the evaluable 21 patients, 1
(4%) was rated class IV, 14 (52%) were class II, and 6 (22%)
were class I. One patient who suffered from severe right
heart failure and received the right ventricular assist device
remained at class IV at the 13 weeks primary endpoint.
Twenty-five patients (76%, 86% of the patients who
survived >30 days) who met eligibility criteria were
discharged from the hospital, with device support. The
median length of time from implant to discharge was 39 days
4.3. Performance outcomes for CE-mark trial
Performance outcomes were evaluated for 33 CE-mark
trial patients at the database closure of June 15, 2007.
The average pump flow rate was maintained over 5 l/min
and the average pump index was over 2.7 l/min/m2 at an
average motor speed of 1700 rpm throughout the support
duration.
Table 5 shows average values for end-organ functions
measured by creatinine, blood urea nitrogen (BUN), total
bilirubin, glutamic pyruvic transaminase (GPT), glutamic
oxaloacetic transaminase (GOT), lactate dehydrogenase
(LDH) and hemolysis parameters at baseline, 4 weeks, 13
weeks, and 6 months. By approximately 2 weeks postimplant, the average values for renal and hepatic functions
decreased to nearly normal ranges for the duration of the
study period. LDH value increased after implantation, then
rapidly decreased during the first postoperative week and
Table 4
Incidence of serious adverse events during support for 33 trial patients.
Serious adverse event
Overall (n = 33) 17.8 P-Y
Initial 11 patients 4.8 P-Y
Last 22 patients 13.0 P-Y
No. of
events
No. of
patients (%)
Event rate
per P-Y
No. of
events
No. of
patients (%)
Event rate
per P-Y
No. of
events
No. of
patients (%)
Event rate
per P-Y
All serious adverse events
Infection; total
Local, non-device related
Driveline
Pocket
Sepsis
92
24
11
6
1
6
28
20
11
5
1
6
(85)
(61)
(33)
(15)
(3)
(18)
5.17
1.35
0.62
0.34
0.06
0.34
34
8
4
2
0
2
10
7
4
1
0
2
(91)
(64)
(36)
(9)
(0)
(18)
7.11
1.67
0.84
0.42
0
0.42
58
16
7
4
1
4
18
13
6
4
1
4
(82)
(59)
(27)
(18)
(5)
(18)
4.45
1.23
0.54
0.31
0.08
0.31
Right heart failure, total
Requiring RVAD
10
1
9 (27)
1 (3)
0.56
0.06
4
0
3 (27)
0 (0)
0.84
0
6
1
6 (27)
1 (5)
0.46
0.08
Neurological dysfunction; total
CVA
TIA
10
5
5
9 (27)
5 (15)
5 (15)
0.56
0.28
0.28
7
5
2
6 (55)
5 (45)
2 (18)
1.46
1.05
0.41
3
0
3
3 (14)
0 (0)
3 (14)
0.23
0
0.23
Ventricular arrhythmia
Renal dysfunction — acute
7
4
7 (21)
4 (12)
0.39
0.23
2
3
2 (18)
3 (27)
0.42
0.63
5
1
5 (23)
1 (5)
0.38
0.08
Bleeding; total
Requiring surgery
8
4
8 (24)
4 (12)
0.45
0.22
2
1
2 (18)
1 (9)
0.42
0.21
6
3
6 (27)
3(14)
0.46
0.23
Respiratory failure
4
4 (12)
0.22
1
1 (9)
0.21
3
3 (14)
0.23
Temporary flow interruption
Hepatic dysfunction
3
2
2 (6)
2 (6)
0.17
0.11
0
1
0 (0)
1 (9)
0
0.21
3
1
2 (9)
1 (5)
0.23
0.08
16
12 (36)
0.90
5
5(45)
1.05
11
8 (36)
0.84
Other; total
P-Y: patient-year, RVAD: right ventricular assist device, CVA: cerebrovascular accident, TIA: transient ischemic attack.
Please cite this article in press as: Morshuis M, et al. European experience of DuraHeartTM magnetically levitated centrifugal left
ventricular assist system. Eur J Cardiothorac Surg (2009), doi:10.1016/j.ejcts.2008.12.033
EJCTS-6818; No of Pages 9
M. Morshuis et al. / European Journal of Cardio-thoracic Surgery xxx (2009) xxx—xxx
7
Table 5
End-organ function and hemolysis during support.
Variables
BUN (mg/dl)
Creatinine (mg/dl)
Total bilirubin (mg/dl)
GOT (U/l)
GPT (U/l)
Free hemoglobin (mg/dl)
LDH (U/l)
Mean SD
Baseline (n = 33)
4 weeks (n = 30)
13 weeks (n = 24)
6 months (n = 15)
38.7 26.0
1.5 0.5
22.0 19.6
1.2 0.4
24.1 21.1
1.3 0.7
27.9 14.5
1.4 0.6
1.4 1.1
89 286
115 429
1.2 2.1
36 22
31 23
0.7 0.4
32 10
28 11
0.6 0.2
39 37
21 14
10.0 10.6
295 120
8.1 5.8
323 96
11.0 10.0
267 73
9.0 7.5
260 77
BUN: blood urea nitrogen, GPT: glutamic pyruvic transaminase, GOT: glutamic oxaloacetic transaminase, LDH: lactate dehydrogenase.
(range: 23—147). No death occurred in an outpatient setting.
Seventeen discharged patients (68%) required rehospitalization due to complications with a median duration of
rehospitalizaiton of 10 days (range: 1—111).
4.4. Adverse event for CE-mark trial
Adverse events were analyzed for 33 CE-mark trial
patients at the time of database closure of June 15, 2007.
All adverse events were monitored and adjudicated by the
CEC (Table 4). The mean duration of support was 197 172
days (range: 17—711, median: 139 days) with a cumulative
support duration of 6500 days. A total of 92 adverse events
were observed in 28 patients (85%) during the support.
The most frequently occurred adverse event was infection. Twenty-four infections occurred in 19 patients (58%).
Eleven of these events (33%) were localized and non-device
related, while 6 (18%) had device related infection (5
driveline and 1 driveline/pocket infection). Six patients
(18%) had sepsis.
Ten events of right heart failure were reported in nine
patients (27%) and one patient required a right ventricular
assist device (RVAD) with the Thoratec PVADTM (Thoratec
Corp., Pleasanton, CA, USA) (3%). The pump flow of RVAD has
been maintained at 10—20% lower than that of LVAD.
Ten neurological events occurred in nine patients (27%),
five were CVA and five were transient ischemic attacks
(TIA). Of these five CVAs, four CVAs (three hemorrhagic and
one ischemic) were determined to be the cause of death.
One intracerebral bleeding following an accidental fall was
resolved without permanent neurological deficit. Five CVAs
were reported in the initial 11 patients, whereas no CVA
was reported in the last 22 patients after optimization of
anticoagulation.
Four patients developed acute renal dysfunction (12%)
perioperatively; however, all events resolved within a few
days.
Eight bleeding events occurred in eight patients (24%).
Four events requiring surgical intervention, two events of
cardiac tamponade and two pocket bleedings, were adjudicated as device-related. One pocket bleeding in the initial 11
patient group was recurrent and was likely related to an
excessive anticoagulation therapy (PTT: 98 s. and INR: 4.15).
Three events of sudden temporary flow interruption
occurred in two patients (6%).
In each case, the pump was subsequently restarted;
however, the decision was made at the study sites to
electively replace the pump. The patients did not experience
any adverse clinical consequence related to the temporary
flow interruption or pump replacement. The returned
pumps were fully functional without any sign of mechanical
failure. The manufacturer identified the root cause of
temporary flow interruption and implemented a manufacturing process change in the motor without design change.
The clinical experience to date has shown no pump malfunctions for patients with improved motors.
There was no device mechanical failure, device thrombosis, hemolysis, peripheral thromboembolism, myocardial
infarction, or chronic renal failure.
5. Discussion and conclusions
In this study, we evaluated the safety and performance of
the DuraHeart LVAS, a 3rd generation rotary pump LVAD
combined with the magnetic levitation technology and a
centrifugal pump. The results demonstrated that the
DuraHeart was safe and performed as intended for mechanical circulatory support for patients at imminent risk of death
due to end-stage left ventricular failure and eligible to
cardiac transplantation. The DuraHeart LVAS provided
effective hemodynamic support for a wide range of body
sizes, fast recovery of end-organ functions, improved
functional status, and a survival benefit to the patients both
inside and outside the hospital.
The adverse event rates per patient-years for major
adverse events during the DuraHeart support were acceptable in comparison with the 1st generation pulsatile [2] and
the 2nd generation axial flow LVADs [6]. The event rate of
bleeding requiring surgery (0.22 per patient-year) was
considerably lower for the DuraHeart than the rate for both,
1st and 2nd generation LVADs (1.47 and 0.78). Driveline or
pocket infection rate was reduced by 90% as compared with
the pulsatile device (0.40 vs 3.49) and comparable to the
small axial flow devices (0.40 vs 0.37).
The rate of neurological events for 33 patients was 50%
less than that of pulsatile devices (0.56 vs 1.11) and slightly
higher than the rate in the axial flow device group (0.56 vs
0.45). A high incidence of fatal intra-cerebral bleeding and
other bleeding complications was observed in the initial 11
patients and the anticoagulation and anti-platelet regimen
was reviewed and discussed by the investigators. The
investigators recognized that the risk of bleeding complication increased with age [13] and the majority (75%) of
Please cite this article in press as: Morshuis M, et al. European experience of DuraHeartTM magnetically levitated centrifugal left
ventricular assist system. Eur J Cardiothorac Surg (2009), doi:10.1016/j.ejcts.2008.12.033
EJCTS-6818; No of Pages 9
8
M. Morshuis et al. / European Journal of Cardio-thoracic Surgery xxx (2009) xxx—xxx
patients who died of CVA with massive intracerebral
bleeding and subdural hematome were older than 60 years.
Such fatal bleeding was likely caused by a combination of
older age and excessive anticoagulation/anti-platelet
therapy, which involved intravenous heparin with a PTT
>90 s and an INR >4.0, and multiple dosages of antiplatelet agents including a combination of clopidogrel,
aspirin and dipyridamole. The investigators agreed to
follow less intensive anticoagulation/anti-platelet regimen
recommended in the protocol. There were no additional
incidences of CVA in the remainder of the study. There was
only 1 late death (178 days) due to sepsis in the last 22
patient group. The Kaplan—Meier analysis of stroke free
survival estimate at 6 months and 1 year was 36% for the
initial 11 and 92% for the last 22 patients (log rank,
p < 0.0001). The mean age for the initial 11 and the last 22
patients was 57 14 and 55 13 years, respectively,
showing no significant difference in age between two
groups (Mann—Whitney test, p = 0.76). After implementing
less intensive anticoagulation, there was no incidence of
stroke for the last 22 patients and the overall neurological
event rate for this patient group was reduced by 50%
compared with the axial flow devices (0.23 vs 0.45). It is
important to keep the anticoagulation/anti-platelet regimen as low as possible, especially for the older patient
cohort in order to avoid fatal hemorrhagic CVA, given the
aging population of transplant candidates in recent years.
No device thrombosis, hemolysis, peripheral thromboembolic events, myocardial infarction, or chronic renal
dysfunction were reported in the present study.
The median time to transplantation was 142 days (range:
43—497), which was 1.5 times longer than that of the most
recent clinical study (97 days) in the United States [6] and 35
patients (51%) are still awaiting heart transplant with a mean
support duration of 317 days (range: 19—1148, median: 216).
At 1 year of support, only 16 patients (38%) have undergone
transplant, while 13 patients (31%) still remain on device
support. In this study, the median age was 58 years (range:
29—74) and 43% of the patients were older than 60 years and
31% were older than 65 years, which is older than the patient
cohort in the most recent clinical study for bridge to
transplantation in the United States (mean age: 50 years) and
some of the ongoing patients have been converted to
destination therapy due to patient preference or delisted
from transplantation according to national transplant law in
each European county.
In conclusion, the initial clinical results demonstrated that
the DuraHeart LVAS is safe and provides an adequate
circulatory support with an acceptable adverse event rate
for the patients eligible for cardiac transplantation. The
device may have a potential for long-term circulatory
support not only as a bridge to cardiac transplantation,
but also for the older patient cohort as a destination therapy.
However, the limitations of the current study include a
limited clinical experience with 68 patients and lack of
direct, randomized comparison with other LVADs or optimal
medical therapy. Further clinical investigation will be
necessary to evaluate that the DuraHeart continues to
provide safe and reliable operation and survival benefit to the
patients with advanced heart failure for bridge to transplant
application as well as for destination therapy.
References
[1] Rose EA, Gelijns AC, Moskowitz AJ, Heitjan DF, Stevenson LW, Dembitsky
W, Long JM, Ascheim DD, Tierny AR, Levitan RG, Watson JT, Meier P. Longterm use of a left ventricular assist device for end-stage heart failure. N
Eng J Med 2001;345:1435—43.
[2] Frazier OH, Rose EA, Oz MC, Dembitsky W, McCarthy PM, Radovancevic B,
Poirier VL, Dasse KA. Multicenter clinical evaluation of the HeartMate
vented electric left ventricular assist system in patients awaiting heart
transplantation. J Thorac Cardiovasc Surg 2001;122:1186—95.
[3] Deng MC, Edwards LB, Hertz MI, Rowe AW, Keck BM, Kormos R, Naftel DC,
Kirklin JK, Taylor DO. Mechanical circulatory support device database of
the International Society for Heart and Long Transplantation: third annual
report — 2005. J Heart Lung Transplant 2005;24:1182—7.
[4] Goldstein DJ. Worldwide experience With the MicroMed DeBakey ventricular assist device as a bridge to transplantation. Circulation
2003;108(Suppl. II):II272—7.
[5] Frazier OH, Myers TJ, Westaby S, Gregoric JD. Clinical experiences with
an implantable, intracardiac, continuous flow circulatory support device:
physiologic implication and their relationship to patient selection. Ann
Thorac Surg 2004;77:133—42.
[6] Miller LW, Pagani FD, Russsell AD, John R, Boyle AJ, Aaronson KD, Conte
JV, Naka Y, Mancini D, Delgado RM, MacGillivray TE, Farrar DJ, Frazier OH.
Use of continuous-flow device in patients awaiting heart transplantation.
N Engl J Med 2007;357:885—96.
[7] Hetzer R, Weng Y, Potapov EV, Pasic M, Drews T, Jurmann M, Henning E,
Mueller J. First experiences with a novel magnetically suspended axial flow
left ventricular assist device. Eur J Cardiothorac Surg 2004;25:961—70.
[8] Esmore DS, Kaye D, Salamansen R, Buckland M, Rowland M, Negri J,
Rowley Y, Woodard J, Begg JR, Ayre P, Rosenfeldt FL. First clinical implant
of the VentrAssist left ventricular assist system as destination therapy for
end-stage heart failure. J Heart Lung Transplant 2005;24:1150—4.
[9] Nojiri C, Kijima T, Maekawa J, Horiuchi K, Kido T, Sugiyama T, Mori T,
Sugiura N. Terumo implantable left ventricular assist system (T-ILVAS).
Results of long-term animal study. ASAIO J 2000;46:117—22.
[10] Saito S, Westaby S, Piggot D, Katsumata, Dudnikov S, Robson D, Catarino
P, Nojiri C. Reliable long-term non-pulsatile circulatory support without
anticoagulation. Eur J Cardiothorac Surg 2001;19:678—83.
[11] Eya K, Tuzum E, Conger J, Chee HK, Byler D, Nojiri C, Frazier OH,
Kadipasaoglu K. Effect of pump flow mode of novel left ventricular assist
device upon end organ perfusion in dogs with doxorubicin induced heart
failure. ASAIO J 2005;51:41—9.
[12] Nishinaka T, Schima H, Roety W, Rajek A, Nojiri C, Wolner E, Wieselthaler
GM. The DuraHeart VAD, a magnetically levitated centrifugal pump — The
university of Vienna bridge to transplant experience. Circ J
2006;70:1421—5.
[13] Rosamonda W, Flegal K, Friday G, Furie K, Go A, Greenlund K, Hasse N, Ho
M, Howard V, Kissela B, Kittner S, Lloyd-Jones D, McDermott M, Meigs J,
Moy C, Nichol G, O’Donnell CJ, Roger V, Rumsfeld J, Sorlie P, Steinberger
J, Thoma T, Wasserthiel-Smoller S, Hong Y. Heart disease and stroke
statistics — 2007 update. Circulation 2007;115:e69—171.
Appendix A. Conference discussion
Dr A. Franco-Cereceda (Stockholm, Sweden): First of all, according to
your inclusion criteria, this device should be implanted in patients eligible for
transplantation, that is transplant candidates. Still, you have one-third of the
patients aged 65 years or older and you have patients being 73 and 74 years old.
Do you actually mean that we should first use a 50,000-Euro LVAD on a 74-yearold and then transplant him? How did you include these patients in the study?
I’m asking you this because it obscures the results with this difference in
age. That’s my first question.
The second question relates to what you have told us here, but also what
you told us at the Techno-College, is that the pumps we have now are getting
better, but we are not getting better in relation to the pumps. We still have the
same types of complications, the cerebro-vascular accidents, the driveline
infections and the right ventricular failure.
What does this mean? Are we getting sloppy or are we using the pumps so
long that we will eventually see these types of complications arise again?
My third question also relates to what we heard earlier, the right
ventricular failure. You have approximately 20% of patients going into right
ventricular failure. It would be interesting to hear what kind of patients these
Please cite this article in press as: Morshuis M, et al. European experience of DuraHeartTM magnetically levitated centrifugal left
ventricular assist system. Eur J Cardiothorac Surg (2009), doi:10.1016/j.ejcts.2008.12.033
EJCTS-6818; No of Pages 9
M. Morshuis et al. / European Journal of Cardio-thoracic Surgery xxx (2009) xxx—xxx
are and how you manage them and what was the outcome. Particularly, is there
any way you can foresee who these patients are?
Dr Morshuis: The first question about age, in Germany we don’t have a real
age limit for transplant. We can also transplant a 72-year-old patient when he
is in a good condition, so we can include them in our protocol. That’s not a
problem. Although, there is, of course, a shortage of donor organs.
And as I pointed out, of course, age is a risk, but we do also support the
older patients when they have a good condition besides the heart.
To the second question, we still see adverse events, of course. Because I think
we are working with a very ill category of patients, and I think partially they bring
their potential trouble already with them before you implant the LVAD.
So I think when you have a survival of 80% after one year, that’s the best you
can reach in this ill category of patients. Because, otherwise, you have to go to a
more healthy category of patient who in fact do not need an assist device. So I
think you can never operate without adverse events in this category of patients.
And also, for example, the drivelines improved, but you have to care very
securely for the driveline, and though you will see some events. So maybe it
would be a good option to make a fully implantable assist device.
9
And to your third question, please help me, the third?
Dr Franco-Cereceda: Right ventricular failure.
Dr Morshuis: Yes. Of course, that’s a difficult point. We try to predict if the
right ventricle will perform well because when we have strong doubts about
the right ventricle, we still implant the Thoratec PVAD because then it’s very
easy to also support the right side.
When we have, unfortunately, a big problem on the right side after
LVAD implantation, then we tend to take the Levitronix when we expect
recovery. But, for example, we had a patient, He was 180 days on an
LVAD with Levitronix on the right side, and then he got a transplant. When
we do not expect recovery, then we use the Thoratec PVAD on the
right side.
So it can be the case that the patient has an LVAD on the left side, for
example, the DuraHeart or the HeartMate, and he has a Thoratec PVAD on the
right side. We don’t like it, but it is a solution for the patient. And they cope, in
fact, quite well with it.
But you can discuss a lot about right ventricular function. It’s difficult to
predict sometimes.
Please cite this article in press as: Morshuis M, et al. European experience of DuraHeartTM magnetically levitated centrifugal left
ventricular assist system. Eur J Cardiothorac Surg (2009), doi:10.1016/j.ejcts.2008.12.033