Download Right Heart Dysfunction After Left Ventricular Assist Device

ADULT CARDIAC
Right Heart Dysfunction After Left Ventricular
Assist Device Implantation: A Comparison of the
Pulsatile HeartMate I and Axial-Flow
HeartMate II Devices
Nishant D. Patel, BA, Eric S. Weiss, MD, Justin Schaffer, MS, Susan L. Ullrich, RN,
Dennis C. Rivard, MA, Ashish S. Shah, MD, Stuart D. Russell, MD, and
John V. Conte, MD
Divisions of Cardiac Surgery and Cardiology, The Johns Hopkins Medical Institutions, Baltimore, Maryland
Background. Right heart dysfunction confers significant morbidity and mortality after left ventricular assist
device implantation and historically occurs in as many
as a third of patients. It is unknown whether newer
axial flow pumps have a different impact on postimplant right heart dysfunction. We compared the incidence of right heart dysfunction after implantation of the
pulsatile HeartMate I (XVE) and the continuous flow
HeartMate II left ventricular assist device.
Methods. We retrospectively reviewed patients who
underwent HeartMate I or HeartMate II implantation
between June 2000 and March 2007. Right heart dysfunction was defined as inotropic/vasodilator support for 14
or more consecutive days or the need for a right ventricular assist device, or both.
Results. Seventy-seven patients underwent HeartMate
implantation; 43 received a HeartMate I and 34 received
a HeartMate II, for a mean left ventricular assist device
support time of 202 and 160 days, respectively. Operative
mortality was lower for HeartMate II patients (28%
versus 15%; p ⴝ 0.26). The HeartMate II patients had
lower preoperative right ventricular stroke work index.
Pulmonary vascular resistance index, right ventricular
stroke work index, and pulmonary and right atrial pres-
sures improved and were similar between groups postoperatively. Overall, right heart dysfunction developed in 35% of HeartMate I patients (15 of 43) and 41%
of HeartMate II patients (14 of 34; p ⴝ 0.63). Fewer
HeartMate II patients (2) than HeartMate I patients (5)
required 7 or more days of epinephrine, whereas more
HeartMate II patients (7) than HeartMate I patients (5)
required 7 or more days of milrinone. Six HeartMate I
and 3 HeartMate II patients required right ventricular
assist device implantation for right heart failure. Survival
was similar (p ⴝ 0.7) between groups at, respectively, 3
(63% versus 62%), 6 (58% versus 58%), and 12 months
(49% versus 48%).
Conclusions. Right heart dysfunction is a persistent
clinical problem after left ventricular assist device
placement. We report the first study comparing the
incidence of right heart dysfunction after HeartMate I
versus HeartMate II implantation. Although the incidence of right heart dysfunction was similar, fewer
HeartMate II patients required right ventricular assist
device placement and fewer required pure inotropic
support for right heart failure.
(Ann Thorac Surg 2008;86:832– 40)
© 2008 by The Society of Thoracic Surgeons
N
but it is donor limited, requires immunosupression therapy, and has many exclusions to candidacy.
Left ventricular assist devices (LVAD) have been
shown to successfully support patients awaiting cardiac
transplantation [4 –12]. The Randomized Evaluation of
Mechanical Assistance for the Treatment of Congestive
Heart Failure (REMATCH) trial showed that LVADs may
improve survival and quality of life for those with severe
end-stage CHF who are not candidates for cardiac transplantation [13, 14]. The REMATCH trial demonstrated a
48% reduction in the risk of death for LVAD recipients
early 5 million Americans suffer from congestive
heart failure (CHF) [1]. With 550,000 new cases diagnosed annually and cost estimates exceeding $10 billion,
CHF continues to be a significant public health concern
[1, 2]. Survival remains poor, with a 1-year mortality rate
greater than 50% for patients with severe CHF [3].
Survival, quality of life, and functional capacity are
limited for patients with CHF despite significant advancements in medical therapy. Currently, cardiac transplantation is the gold standard for the treatment of CHF,
Accepted for publication May 5, 2008.
Presented at the Forty-fourth Annual Meeting of The Society of Thoracic
Surgeons, Fort Lauderdale, FL, Jan 28 –30, 2008.
Address correspondence to Dr Conte, Division of Cardiac Surgery, Johns
Hopkins Medical Institutions, Blalock 618, 600 N Wolfe St, Baltimore,
MD 21287; e-mail: [email protected].
© 2008 by The Society of Thoracic Surgeons
Published by Elsevier Inc
Drs Conte and Russell disclose that they have a financial relationship with Thoratec Corp.
0003-4975/08/$34.00
doi:10.1016/j.athoracsur.2008.05.016
PATEL ET AL
HEARTMATE II VERSUS I LVAD
833
Table 1. Baseline Characteristics
Abbreviations
CHF
CVP
LVAD
mPAP
NYHA
PCWP
and Acronyms
⫽ congestive heart failure
⫽ central venous pressure
⫽ left ventricular assist device
⫽ mean pulmonary artery pressure
⫽ New York heart association
⫽ pulmonary capillary wedge
pressure
REMATCH ⫽ Randomized Evaluation of
Mechanical Assistance for the
Treatment of Congestive Heart
Failure
RHD
⫽ right heart dysfunction
RVAD
⫽ right ventricular assist device
RVSWI
⫽ right ventricular stroke work index
versus patients receiving optimal medical therapy despite using an LVAD with limited durability.
Right heart dysfunction (RHD) is a concern after LVAD
implantation and occurs historically in as many as one
third of patients [15]. First-generation pulsatile devices,
such as the HeartMate XVE, have been causally related to
the incidence of RHD owing to the mechanical effects of
left ventricular unloading [15]. The impact of recently
developed axial flow [16 –19] LVADs on right heart function is unknown. The HeartMate II LVAD (Thoratec
HeartMate II
(n ⫽ 34)
HeartMate I
(n ⫽ 43)
p
Value
51.4 ⫾ 14.5
27 (79.4)
48.7 ⫾ 13.2
36 (83.7)
0.40
0.62
16 (47.1)
15 (34.9)
0.35
13 (38.2)
15 (34.9)
0.81
0 (0)
2 (5.9)
2 (4.7)
4 (9.3)
0.50
0.69
0 (0)
2 (4.7)
0.50
0 (0)
1 (2.9)
1 (2.3)
1 (2.3)
1.00
1.00
1 (2.9)
0 (0)
0 (0)
1 (2.3)
1 (2.3)
1 (2.3)
1.00
1.00
1.00
1 (2.9)
9 (26.5)
16 (47.1)
19 (55.9)
17 (50.0)
8 (24.2)
13 (38.2)
0 (0)
13 (30.2)
23 (53.5)
24 (55.8)
17 (39.5)
8 (21.6)
18 (41.9)
0.44
0.71
0.57
0.99
0.35
0.79
0.74
11 (32.4)
20 (46.5)
0.20
Age (years)
Male sex (%)
Indications
Idiopathic
cardiomyopathy (%)
Ischemic
cardiomyopathy (%)
Valvular disease (%)
Familial
cardiomyopathy (%)
Postpartum
cardiomyopathy (%)
Radiation induced (%)
Adriamycin
cardiomopathy (%)
Myocarditis (%)
Rheumatic disease (%)
Accelerated
atherosclerosis
post-OHT (%)
Congenital (%)
Diabetes mellitus (%)
Hyperlipidemia (%)
Hypertension (%)
Smoking (%)
Cardiogenic shock (%)
Previous myocardial
infarction (%)
Preoperative intra-aortic
balloon pump (%)
OHT ⫽ orthotopic heart transplantation.
Corporation, Pleasanton, California) is a promising new
axial-flow pump that is currently in clinical trials in the
United States (Fig 1). Complete left ventricular unloading
is avoided with this device and other second-generation
axial flow pumps. Our sense and that of other investigators is that there is less RHD with the axial flow pumps
owing to maintenance of left ventricular end-diastolic
volume, maintained septal position, and preservation
of right ventricular mechanics. We began using the
HeartMate II in January of 2005 and sought to compare
the incidence of RHD and right heart failure after implantation with the HeartMate II versus HeartMate I
LVAD.
Material and Methods
Study Design
Fig 1. (Left) The HeartMate II and (right) the HeartMate I (XVE)
left ventricular assist devices.
We conducted a retrospective review of all patients who
underwent either HeartMate II or HeartMate I (XVE)
LVAD implantation at our institution from June 2000 to
March 2007 after Institutional Review Board approval.
Individual waiver for consent was granted. For analysis,
patients were grouped into those who underwent
ADULT CARDIAC
Ann Thorac Surg
2008;86:832– 40
834
ADULT CARDIAC
PATEL ET AL
HEARTMATE II VERSUS I LVAD
Ann Thorac Surg
2008;86:832– 40
Table 2. Baseline Hemodynamics, Renal and Hepatic
Function, and Hematologic Data
HeartMate II HeartMate I
p
(n ⫽ 34)
(n ⫽ 43)
Value
Hemodynamics
Heart rate (beats/min)
Systolic blood pressure
(mm Hg)
Ejection fraction (%)
PCWP (mm Hg)
mPAP (mm Hg)
CVP (mm Hg)
PVRI
(dynes/sec*cm⫺5/m2)
Cardiac index
(L · min⫺1 · m⫺2)
RVSWI (g*m/m2)
Renal and hepatic
function
Blood urea nitrogen
(mg/dL)
Creatinine (mg/dL)
Total bilirubin (mg/dL)
Alanine
aminotransferase
(IU)
Aspartate
aminotransferase
(IU)
90.1 ⫾ 20.3
100.1 ⫾ 12.4
86.6 ⫾ 20.0
99.2 ⫾ 16.6
0.45
0.80
14.2 ⫾ 6.7
25.3 ⫾ 10.5
34.2 ⫾ 12.2
12.2 ⫾ 6.5
433 ⫾ 269
12.6 ⫾ 6.2
25.5 ⫾ 7.6
35.9 ⫾ 9.2
12.4 ⫾ 5.9
512 ⫾ 435
0.28
0.93
0.52
0.87
0.38
1.8 ⫾ 0.5
1.9 ⫾ 0.47
0.47
5.7 ⫾ 2.3
8.2 ⫾ 3.9
0.002
42.4 ⫾ 24.4
37.0 ⫾ 21.1
0.30
1.7 ⫾ 0.8
1.5 ⫾ 1.1
92.1 ⫾ 221.3
1.7 ⫾ 0.9
1.9 ⫾ 1.5
82.3 ⫾ 96.0
0.87
0.20
0.79
48.9 ⫾ 63.8
72.1 ⫾ 90.6
0.21
CVP ⫽ central venous pressure;
mPAP ⫽ mean pulmonary artery
pressure;
PCWP ⫽ pulmonary capillary wedge pressure;
PVRI ⫽
pulmonary vascular resistance index;
RVSWI ⫽ right ventricular
stroke work index.
HeartMate II implantation versus those who had
HeartMate I implantation. Relevant baseline, operative,
and postoperative data were collected. Hemodynamic
Fig 2. Preoperative and postoperative hemodynamic data for
HeartMate I (solid lines) and HeartMate II (dashed lines). (CVP ⫽
central venous pressure; mPAP ⫽ mean pulmonary artery pressure;
PCWP ⫽ pulmonary capillary wedge pressure.)
data included pulmonary capillary wedge pressure
(PCWP), mean pulmonary artery pressure (mPAP), central venous pressure (CVP), cardiac index, pulmonary
vascular resistance index, and right ventricular stroke
work index (RVSWI) collected preoperatively and 1, 3,
and 6 months after LVAD implantation. Study endpoints
included survival, the incidence of right heart dysfunction, defined as the need for inotropes or vasodilators for
14 or more consecutive days, or the need for a right
ventricular assist device (RVAD).
Statistical Analysis
Statistical analyses were performed using SPSS 12.0
software (SPSS, Chicago, Illinois). All data are presented
as HeartMate II versus HeartMate I and as mean ⫾ SD
unless otherwise noted. Fisher’s exact test and t test were
used for qualitative and quantitative variable analyses,
Table 3. Preoperative and Postoperative Hemodynamic Indexes
HeartMate I hemodynamics
PCWP (mm Hg)
mPAP (mm Hg)
CVP (mm Hg)
PVRI (dynes/sec*cm⫺5/m2)
Cardiac index (L · min⫺1 · m⫺2)
RVSWI (g*m/m2)
HeartMate II hemodynamics
PCWP (mm Hg)
mPAP (mm Hg)
CVP (mm Hg)
PVRI (dynes/sec*cm⫺5/m2)
Cardiac index (L · min⫺1 · m⫺2)
RVSWI (g*m/m2)
a
Preoperative
1 Month
3 Months
6 Months
25.5 ⫾ 7.6
35.9 ⫾ 9.2
12.4 ⫾ 5.9
512 ⫾ 435
1.9 ⫾ 0.47
8.2 ⫾ 3.9
10.2 ⫾ 6.6a
20.9 ⫾ 7.1a
9.9 ⫾ 5.7
334 ⫾ 184
2.8 ⫾ 0.8a
4.8 ⫾ 2.7a
13.4 ⫾ 7.6a
24.0 ⫾ 7.8a
10.8 ⫾ 8.1
303 ⫾ 74
2.9 ⫾ 0.8a
5.6 ⫾ 2.1
10.6 ⫾ 3.0a
21.5 ⫾ 4.2a
8.6 ⫾ 4.0
322 ⫾ 66
2.7 ⫾ 0.7a
5.6 ⫾ 3.1
25.3 ⫾ 10.5
34.2 ⫾ 12.2
12.2 ⫾ 6.5
433 ⫾ 269
1.8 ⫾ 0.5
5.7 ⫾ 2.3
8.6 ⫾ 5.6a
19.0 ⫾ 6.8a
8.8 ⫾ 6.5
290 ⫾ 131a
3.0 ⫾ 0.5a
5.4 ⫾ 2.6
11.9 ⫾ 4.7a
22.7 ⫾ 7.5a
9.7 ⫾ 4.9
345 ⫾ 211
2.8 ⫾ 0.5a
5.4 ⫾ 1.9
11.2 ⫾ 8.6a
20.0 ⫾ 9.3a
9.7 ⫾ 8.2
276 ⫾ 152
2.69 ⫾ 0.7a
4.2 ⫾ 2.4
p ⬍ 0.05 versus preoperative.
CVP ⫽ central venous pressure;
mPAP ⫽ mean pulmonary artery pressure;
PCWP ⫽ pulmonary capillary wedge pressure;
pulmonary vascular resistance index;
RVSWI ⫽ right ventricular stroke work index.
PVRI ⫽
PATEL ET AL
HEARTMATE II VERSUS I LVAD
Table 4. Postoperative Consecutive Days Requiring Inotropes
for HeartMate I and HeartMate II Patients
Number of Patients
HeartMate II
HeartMate I
1
2
0
5
1
7
0
5
Epinephrine
⬎ 0.05 ␮g · kg⫺1 · min⫺1
ⱖ 14 days
ⱖ 7 days
Milrinone
⬎ 0.50 ␮g · kg⫺1 · min⫺1
ⱖ 14 days
ⱖ 7 days
respectively. Kaplan-Meier and log-rank analyses were
performed to compare survival for HeartMate II versus
HeartMate I patients. We conducted Cox proportional
hazards regression modeling to assess for predictors of
mortality and logistic regression analysis to assess for
predictors of right heart dysfunction.
Operative Technique
Patients receiving the HeartMate II LVAD underwent
median sternotomy. The outflow graft was anastomosed
to the ascending aorta without bypass whenever possible. After the institution of cardiopulmonary bypass, an
apical ventriculotomy was made and the inflow cannula
was placed in the left ventricular apex. The pump was
then activated; after deairing, patients were weaned from
cardiopulmonary bypass. Our patients routinely come off
cardiopulmonary bypass on epinephrine, milrinone, and
inhaled nitric oxide in various combinations. The continuation of inotropes was made based on clinical
interpretation of the hemodynamic data. Those patients requiring protracted inotropes typically had inotrope-dependent LVAD flow as well as echocardiographic evidence of right ventricular dysfunction.
The HeartMate I LVAD is implanted in a similar
fashion. HeartMate I patients were weaned off cardiopul-
835
monary bypass in a fixed rate mode, as recommended by
the manufacturer, and once stable off cardiopulmonary
bypass, switched to automatic mode. Patients with severe
tricuspid regurgitation underwent concomitant tricuspid
annuloplasty. This decision was made based on the
preoperative echocardiogram in most cases. In some
patients, intraoperative transesophageal echocardiograms that showed significant worsening of tricuspid
regurgitation prompted repair. Tricuspid repair was routinely performed before apical ventriculotomy.
Results
During the study period, 77 patients underwent implantation with a HeartMate LVAD. Since January 2005, 34 patients have received a HeartMate II LVAD. The remaining
43 patients in this study received a HeartMate I LVAD
between June 2000 and October 2006. Mean LVAD support
time was 160.5 ⫾ 173.4 days for the HeartMate II group and
202.3 ⫾ 258.9 days for the HeartMate I group. Baseline
clinical characteristics were similar between groups and are
shown in Table 1. An ischemic cause of CHF occurred in
approximately one third of patients in each group (p ⫽ 1.00).
All HeartMate II patients and 97.6% of HeartMate I patients
(42 of 43) were in preoperative New York Heart Association
(NYHA) class IV.
Baseline hemodynamic indices, renal and hepatic
function, and hematologic data are shown in Table 2 and
were similar between HeartMate II and I patients.
Preoperative RVSWI was significantly lower for the
HeartMate II group (5.7 versus 8.2 g*m/m2; p ⫽ 0.002),
indicating worse right heart function in the HeartMate II
versus HeartMate I group.
Operative Data
Mean cardiopulmonary bypass time was 76.7 ⫾ 34.9
minutes for the HeartMate II group and 96.2 ⫾ 37.5
minutes for the HeartMate I group (p ⫽ 0.04). Twentynine percent of HeartMate II patients (10 of 34) and 4.7%
of HeartMate I patients (2 of 43) underwent concomitant
tricuspid valve annuloplasty (p ⫽ 0.004).
Table 5. Preoperative and Postoperative Renal and Hepatic Data
HeartMate I renal/hepatic function
Blood urea nitrogen (mg/dL)
Creatinine (mg/dL)
Total bilirubin (mg/dL)
Alanine aminotransferase (IU)
Aspartate aminotransferase (IU)
HeartMate II renal/hepatic function
Blood urea nitrogen (mg/dL)
Creatinine (mg/dL)
Total bilirubin (mg/dL)
Alanine aminotransferase (IU)
Aspartate aminotransferase (IU)
a
p ⬍ 0.05 versus preoperative.
Preoperative
1 Month
3 Months
6 Months
37.0 ⫾ 21.1
1.7 ⫾ 0.9
1.9 ⫾ 1.5
82.3 ⫾ 96.0
72.1 ⫾ 90.6
23.7 ⫾ 21.1a
1.4 ⫾ 0.9
2.9 ⫾ 7.21
30.8 ⫾ 40.8a
50.3 ⫾ 68.8
18.6 ⫾ 9.5a
1.2 ⫾ 0.5a
1.0 ⫾ 1.8a
21.1 ⫾ 13.4a
28.0 ⫾ 16.2a
19.2 ⫾ 7.3a
1.1 ⫾ 0.4a
0.5 ⫾ 0.3a
23.6 ⫾ 17.0a
33.7 ⫾ 28.6
42.4 ⫾ 24.4
1.7 ⫾ 0.8
1.5 ⫾ 1.1
92.1 ⫾ 221.3
48.9 ⫾ 63.8
27.1 ⫾ 17.1a
1.4 ⫾ 0.8
2.2 ⫾ 3.2
70.1 ⫾ 165.6
101.8 ⫾ 230.5
22.6 ⫾ 15.1a
1.1 ⫾ 0.5a
1.1 ⫾ 1.1
38.8 ⫾ 43.0
55.4 ⫾ 90.2
25.5 ⫾ 13.4a
1.3 ⫾ 0.7
0.8 ⫾ 0.5a
25.6 ⫾ 14.4
31.1 ⫾ 20.0
ADULT CARDIAC
Ann Thorac Surg
2008;86:832– 40
836
ADULT CARDIAC
PATEL ET AL
HEARTMATE II VERSUS I LVAD
Ann Thorac Surg
2008;86:832– 40
Fig 3. Preoperative and postoperative serum creatinine and total
bilirubin for HeartMate I (solid lines) and HeartMate II (dashed
lines).
Hemodynamics
Changes in hemodynamic indices while on mechanical
circulatory support with either the HeartMate II or
HeartMate I LVAD are shown in Table 3. Mean PCWP,
mPAP, and cardiac index significantly improved postoperatively for the HeartMate II group, whereas the mean
CVP and pulmonary vascular resistance index trended
toward improvement. For the HeartMate I group, mean
PCWP, mPAP, CVP, and cardiac index significantly improved postoperatively; pulmonary vascular resistance
index trended toward improvement, but did not reach
statistical significance. When comparing HeartMate II
and HeartMate I patients, both groups had similar postoperative PCWP, mPAP, and CVP at 1, 3, and 6 months
postoperatively (Fig 2). Postoperative RVSWI was also
similar between groups at 6 months.
Inotropic/Vasodilator Requirement and RHD
Right heart dysfunction, defined as 14 consecutive days
of inotropic or vasodilator support or the need for a right
ventricular assist device (RVAD), was similar between
groups (41.2% versus 34.9%; p ⫽ 0.63). Three HeartMate
Table 6. Postoperative Data
Bleeding requiring
reoperation (%)
Packed red blood cells
(units)a
Fresh frozen plasma
(units)a
Platelets (units)a
Device pocket
infection (%)
Driveline infection (%)
Sepsis (%)
a
HeartMate II
(n ⫽ 34)
HeartMate I
(n ⫽ 43)
p
Value
1 (2.9)
10 (23.3)
0.01
7.3 ⫾ 5.1
9.9 ⫾ 5.6
0.03
7.8 ⫾ 5.0
10.5 ⫾ 6.5
0.04
2.0 ⫾ 1.4
1 (2.9)
5.7 ⫾ 6.4
7 (16.3)
0.001
0.07
10 (29.4)
3 (8.8)
15 (34.9)
9 (20.9)
0.63
0.20
Intraoperative and postoperative.
Fig 4. Kaplan-Meier survival, including hospital deaths, for
HeartMate I (HMI [solid line]) and HeartMate II (HMII [dashed
line]) patients.
II patients and 6 HeartMate I patients required a RVAD
for right heart failure (p ⫽ 0.72).
We assessed postoperative inotropic and vasodilator
requirements for each group, including the number of
consecutive days on epinephrine and milrinone, and
found no differences in the total amount of inotropes and
vasodilators used. The HeartMate II group did have
fewer patients requiring 7 or more consecutive days of
epinephrine, although the HeartMate II group had more
patients requiring 7 or more days of milrinone (Table 4).
Renal and Hepatic Function
Serum markers of renal and hepatic function significantly
decreased for both groups postoperatively (Table 5).
Serum creatinine, total bilirubin, alanine aminotransferase, and aspartate aminotransferase levels trended toward improvement in the HeartMate II group, but were
not quite statistically significant. Serum creatinine, blood
urea nitrogen, alanine aminotransferase, and aspartate
aminotransferase levels did significantly decrease postoperatively in the HeartMate I group (Table 5). All serum
markers were similar at 1, 3, and 6 months postoperatively (Fig 3). Blood urea nitrogen (p ⫽ 0.27), alanine
aminotransferase (p ⫽ 0.83), and aspartate aminotransferase (p ⫽ 0.73) were also similar between groups at 6
months postoperatively.
Length-of-Stay, Complications, and Survival
Total length of stay was 58.4 ⫾ 37.3 days for the HeartMate
II group and 63.4 ⫾ 56.8 days for the HeartMate I group (p ⫽
0.66). Postoperative length of stay was 48.8 ⫾ 32.7 days for
the HeartMate II group and 51.0 ⫾ 54.3 days for the
HeartMate I group (p ⫽ 0.82). Postoperative data are listed
in Table 6. HeartMate II patients required less blood replacement therapy and fewer required reoperation for
bleeding than HeartMate I patients.
Thirty-day mortality was 25.3% (5 of 33) and 27.9% (12
of 43) for the HeartMate II and HeartMate I groups,
respectively (p ⫽ 0.15). Late mortality was 27.3% (9 of 33)
for the HeartMate II group and 32.6% (14 of 43) for the
HeartMate I group, which was not a statistically significant difference (p ⫽ 0.61). As shown in Figure 4, Kaplan-
Meier survival, including hospital deaths, was 50% at 1
year for both groups; 2-year survival was 44% and 48%
for the HeartMate I and II groups, respectively (p ⫽ 0.70).
When analyzing all patients in this series for risk
factors for mortality, increasing age (hazard ratio [HR]
1.07; 95% confidence interval [CI]: 1.02 to 1.12; p ⫽ 0.004)
and lower preoperative RVSWI (HR 0.75; 95% CI: 0.57 to
0.98; p ⫽ 0.03) were associated with mortality. The lone
marker for right heart dysfunction was preoperative
intra-aortic balloon pump counterpulsation (odds ratio
[OR] 6.21; 95% CI: 1.01-38.0; p ⫽ 0.04). Variables included
in the analyses were age, sex, race, type of HeartMate
device, preoperative inotropes, cause of CHF, diabetes
mellitus, cardiogenic shock, chronic obstructive pulmonary disease, previous myocardial infarction, body mass
index, preoperative intra-aortic balloon pump, preoperative creatinine, preoperative PCWP, preoperative
mPAP, preoperative cardiac index, preoperative pulmonary vascular resistance index, preoperative CVP, and
preoperative RVSWI. The type of HeartMate device was
not associated with mortality (HR 0.3; 95% CI: 0.1 to 1.0;
p ⫽ 0.06) or right heart dysfunction (OR 1.3; 95% CI: 0.36
to 5.12; p ⫽ 0.64) in our models.
Comment
Many studies have successfully shown that LVADs are
an excellent option for patients with end-stage CHF and
for those awaiting cardiac transplantation, with significant improvements in survival and quality of life [13, 14,
20 –23]. The REMATCH trial has shown that LVAD therapy improves survival in end-stage heart failure patients
who are not transplant candidates compared with medically managed patients [13, 14]. The two most frequent
causes of death in the LVAD group were sepsis and
device failure [24]. Nearly 50% of the device malfunctions
involved implantable components, the most common of
which was inflow conduit malfunctions resulting in inflow valve incompetence. The newer axial-flow LVADs
have been designed in an effort to minimize operative
risk, improve durability, and lower the risk of devicerelated adverse events, in part by reducing the number of
moving parts in the device [16 –19].
The HeartMate II, developed by Thoratec Corporation,
is a continuous, axial-flow LVAD with a spinning rotor as
its lone moving part, an inflow cannula with a sintered
titanium surface, an impeller powered by an electromagnetic motor, and a single driveline that exits the abdomen
[18, 19, 25–28]. Early results have been excellent in the
bridge to transplant population [28].
Right heart dysfunction is a concern after LVAD implantation and contributes significantly to postoperative
morbidity and mortality. In a study of 108 patients
undergoing HeartMate I implantation, Dang and colleagues [15] reported that 38.9%, or 42 patients, had right
heart failure postoperatively. Of the 42 patients with right
heart failure, 14 required a RVAD. The authors found
that female patients were more likely to develop right
heart failure than male patients, and patients with right
heart failure had a higher early mortality rate, longer
PATEL ET AL
HEARTMATE II VERSUS I LVAD
837
intensive care unit length of stay, higher rates of reoperation for bleeding, and a greater incidence of renal
failure than those who did not have right heart failure.
Furthermore, the authors found that intraoperative CVP
was a significant predictor of postoperative right heart
failure.
Many theories as to the etiology of RHD have been
proposed [29 –33]. Experimental models have been used
to try and discern the genesis of LVAD-associated RHD.
Omoto and colleagues [29] found a Frank-Starling relation for the right ventricular free wall in an isolated right
heart canine model. The authors demonstrated a linear
relationship between right ventricular systolic peak pressure and end-diastolic length or end-diastolic pressure.
Moon and coworkers [30] have shown a global decrease
in right ventricular systolic function with a concomitant
increase in right ventricular end-diastolic volume to
maintain right ventricular output. Farrar and colleagues
[31] in an early work found reduced right ventricular
contraction in normal dogs during LVAD support. Elbeery and colleagues [32], on the other hand, observed
that right ventricular function during LVAD support was
maintained and the systolic interaction with the left
ventricle was of minimal consequence in the maintenance of right heart function. Finally, an earlier work
from Miyamoto and colleagues [33] showed that increases in the amount of left ventricular unloading led to
incremental decreases in the derivative of right ventricular pressure (right ventricular dP/dt), effectively reducing right ventricular function.
A valid criticism of these experimental studies is that
they are performed in normal hearts and not hearts in
animals with CHF. The finding of increased right ventricular end-diastolic pressure experimentally, however,
does correlate with what is seen clinically in those hearts
with RHD after LVAD implantation. They are enlarged,
hypocontractile, and need an increased preload to maintain adequate LVAD flows. Many of the LVAD models
used in these animal experiments are pulsatile pumps,
which do not necessarily reflect what occurs with continuous flow devices.
In our study, 14 HeartMate II patients and 15
HeartMate I patients met our definition of right heart
dysfunction postoperatively. The most significant finding
was that 3 HeartMate II and 6 HeartMate I patients underwent RVAD placement for right heart failure. While the
need for RVAD support was lower in the HeartMate II
group, this did not reach statistical significance. The
length of inotrope and vasodilator use is a much softer
endpoint. Our more recent tendency to leave patients on
vasodilators until diuresed may bias the data. The
HeartMate II patients did require less pure inotropic
support and more vasodilators than HeartMate I patients. A larger patient cohort and longer follow-up may
demonstrate true differences in the need for RVAD and
inotropic support after HeartMate II implantation.
The HeartMate I device mechanically unloads the left
ventricle, which may result in bowing of the interventricular septum away from the right ventricle. Bulging of the
septum into the left ventricle may reduce the efficiency of
ADULT CARDIAC
Ann Thorac Surg
2008;86:832– 40
838
ADULT CARDIAC
PATEL ET AL
HEARTMATE II VERSUS I LVAD
Fig 5. Echocardiogram showing septal bowing into the left ventricle
after HeartMate XVE implantation.
right ventricular contraction by destabilizing the fulcrum
upon which the right ventricle contracts (Fig 5) [15].
Moreover, the right ventricle may receive a venous return beyond its capacity owing to the LVAD’s effective
forward flow through the systemic circulation, resulting
in right ventricular dysfunction [34]. Maintaining the
septal midline position requires maintenance of some
left ventricular volume. This results in less than maximal
LVAD flow, which prevents overcirculation that could
overwhelm the functional capacity of the right ventricle.
In our early experience with the HeartMate II LVAD,
we became more aggressive with performing a concomitant tricuspid annuloplasty, and our current policy is to
repair the valve in patients with moderate or worse
tricuspid regurgitation. That may be important, as experimental evidence suggests diminished right ventricular
function requires an increased right ventricular enddiastolic pressure to maintain comparable forward flow.
This increased right ventricular end-diastolic pressure
and right ventricular end-diastolic dimension can cause
tricuspid valve chordal tethering and increase tricuspid
regurgitation. Tricuspid valve repair can correct this
abnormality and help maintain flow.
Our early clinical impression is that there is less
RHD after HeartMate II implantation. We do find that
HeartMate II patients come off bypass easier, require
lower peak doses of inotropes, and recover relatively
faster than their HeartMate I counterparts. When coming
off cardiopulmonary bypass, we monitor the position of
the septum and attempt to maintain enough left ventricular volume to keep the septum in a midline position,
which may lead to less RHD. On logistic regression
analysis, the only predictor of RHD was preoperative
intra-aortic balloon pump counterpulsation, which demonstrates that preoperative acuity is a predictor of right
heart dysfunction. Although preoperative hemodynamic
indices were not predictive of RHD in our model, we
agree with Dang and colleagues [15] that perhaps the
best predictor of postimplant RHD is preoperative RHD.
The implications of continuous versus pulsatile blood
flow are currently unknown and have been a matter of
significant debate [35]. While some argued that pulsepressure is necessary for maintaining the integrity of the
Ann Thorac Surg
2008;86:832– 40
circulation and end-organ function, others argue that
nonpulsatile flow allows for normal end-organ function.
Many have used animal studies of nonpulsatile flow to
address the issue [35– 41], but few studies are available
that describe outcomes in humans supported by nonpulsatile LVADs. Our early results indicate that renal and
hepatic function after HeartMate II implantation is similar to the HeartMate I group at 1, 3, and 6 months after
surgery.
Recent years have seen considerable advancement in
axial-flow LVADs with the development of the Jarvik
2000 (Jarvik Heart, New York, New York), the MicroMed
DeBakey LVAD (MicroMed, Houston, Texas), the
HeartMate II LVAD, and others. We have shown that the
HeartMate II continuous, axial-flow LVAD adequately
supports the systemic circulation and arguably demonstrates a lower incidence of right heart dysfunction.
Nevertheless, longer follow-up is necessary to determine
if the HeartMate II LVAD can continue to minimize right
heart dysfunction and lower the need for RVAD support
versus older generation pulsatile devices.
This study was supported in part by the Mildred and Carmont
Blitz Cardiac Research Fund. Doctor Weiss is an Irene Piccinini
Investigator in Cardiac Surgery.
References
1. 2001 Heart and stroke statistical update. Dallas: American
Heart Association, 2000.
2. O’Connell JB, Birstow MR. Economic impact of heart failure
in the United States: time for a different approach. J Heart
Lung Transplant 1994;13(Suppl):107–12.
3. Califf RM, Adams KF, McKenna WJ, et al. A randomized
controlled trial of epoprostenol therapy for severe congestive heart failure: the Flolan International Randomized Survival Trial (FIRST). Am Heart J 1997;134:44 –54.
4. Goldstein D, Oz M, Rose E. Implantable left ventricular
assist devices. N Engl J Med 1998;339:1522–33.
5. Frazier O, Rose E, Macmanus Q, et al. Multicenter clinical
evaluation of the HeartMate 1000 IP left ventricular assist
device. Ann Thorac Surg 1992;53:1080 –90.
6. Pennington D, McBride L, Peigh P, et al. Eight years’
experience with bridging to cardiac transplantation. J Thorac
Cardiovasc Surg 1994;107:472– 81.
7. Korfer R, El-Banayosy A, Arusoglu L, et al. Single-center
experience with the Thoratec ventricular assist device. J Thorac Cardiovasc Surg 2000;119:596 – 600.
8. McBride LR, Naunheim KS, Fiore AC, et al. Clinical experience with 111 Thoratec ventricular assist devices. Ann Thorac Surg 1999;67:1233–9.
9. Portner P, Oyer P, Pennington D, et al. Implantable electrical
ventricular assist system: bridge to transplantation and the
future. Ann Thorac Surg 1989;47:142–50.
10. Moskowitz A, Weinberg A, Oz M, Williams D. Quality of life
with an implanted left ventricular assist device. Ann Thorac
Surg 1997;64:1764 –9.
11. Catanese K, Goldstein D, Williams D, et al. Outpatient left
ventricular assist device support: a destination rather than a
bridge. Ann Thorac Surg 1996;61:646 –53.
12. Mehta S, Aufiero T, Pae W, et al. Combined registry for the
clinical use of mechanical ventricular assist pumps and the
total artificial heart in conjunction with heart transplantation: sixth official report. J Heart Lung Transplant 1994;14:
585–93.
13. Rose E, Gelijns A, Moskowitz A, et al. Long-term mechanical
left ventricular assistance for end-stage heart failure. N Engl
J Med 2001;345:1435– 43.
14. Park SJ, Tector A, Piccioni W, et al. Left ventricular assist
devices as destination therapy: a new look at survival.
J Thorac Cardiovasc Surg 2005;129:9 –17.
15. Dang NC, Topkara VK, Mercando M, et al. Right heart
failure after left ventricular assist device implantation in
patients with chronic congestive heart failure. J Heart Lung
Transplant 2006;25:1– 6.
16. Frazier OH, Myers TJ, Westaby S, Gregoric ID. Clinical
experience with an implantable, intracardiac, continuous
flow circulatory support device: physiologic implications
and their relationship to patient selection. Ann Thorac Surg
2004;77:133– 42.
17. Noon GP, Morley DL, Irwin S, Abdelsayed SV, Benkowski
RJ, Lynch BE. Clinical experience with the MicroMed DeBakey ventricular assist device. Ann Thorac Surg 2001;
71(Suppl):133– 8.
18. Griffith BP, Kormos RL, Borovetz HS, et al. HeartMate II left
ventricular assist system: from concept to first clinical use.
Ann Thorac Surg 2001;71(Suppl):116 –20.
19. Frazier OH, Delgado RM, Kar B, Patel V, Gregoric ID, Myers
TJ. First clinical use of the redesigned HeartMate II left
ventricular assist system in the United States. Tex Heart Inst
J 2004;31:157–9.
20. Grady KL, Meyer PM, Dressler D, et al. Longitudinal change
in quality of life and impact on survival after left ventricular
assist device implantation. Ann Thorac Surg 2004;77:1321–7.
21. Dang NC, Topkara VK, Kim BT, et al. Clinical outcomes in
patients with chronic congestive heart failure who undergo
left ventricular assist device implantation. J Thorac Cardiovasc Surg 2005;130:1302–9.
22. Frazier OH, Rose EA, Oz MC, et al. Multicenter clinical
evaluation of the HeartMate vented electrical left ventricular
assist system in patients awaiting heart transplantation.
J Thorac Cardiovasc Surg 2001;122:1186 –95.
23. Koul B, Solem JO, Steen S, Casimir-Ahn H, Granfeldt H,
Lonn UJ. HeartMate left ventricular assist device as bridge to
heart transplantation. Ann Thorac Surg 1998;65:1625–30.
24. Dembitsky WP, Tector AJ, Park S, et al. Left ventricular assist
device performance with long-term circulatory support: lessons from the REMATCH trial. Ann Thorac Surg 2004;78:
2123–30.
25. Long JW. Advanced mechanical circulatory support with the
HeartMate left ventricular assist device in the year 2000. Ann
Thorac Surg 2001;71(Suppl):176 – 82.
26. Burke DJ, Burke E, Parsaie F, et al. The HeartMate II: design
and development of a fully sealed axial flow left ventricular
assist system. Artif Organs 2001;25:380 –5.
PATEL ET AL
HEARTMATE II VERSUS I LVAD
839
27. Butler K, Thomas D, Taylor L, et al. The HeartMate II axial
flow LVAS: journey toward clinical trial. ASAIO J 2000;46:
195.
28. Miller LW, Pagani FD, Russell SD, et al. Use of a continuousflow device in patients awaiting heart transplantation.
N Engl J Med 2007;357:885–96.
29. Omoto T, Tanabe H, LaRia PJ, Guererro J, Vlahakes GJ. Right
ventricular performance during left ventricular unloading
conditions: the contribution of the right ventricular free wall.
Thorac Cardiov Surg 2002;50:16 –20.
30. Moon MR, Castro LJ, De Anda A, et al. Right ventricular
assistance in closed chest dogs. Ann Thorac Surg 1993;56:
54 – 67.
31. Farrar DJ, Compton PG, Dajee H, Fonger JD, Hill JD. Right
heart function during left heart assist and the effects of
volume unloading in a canine preoparation. Circulation
1984;70:708 –16.
32. Elbeery JR, Owen CH, Savitt MA, et al. Effects of the left
ventricular assist device on right heart function. J Thorac
Cardiovasc Surg 1990;99:809 –16.
33. Miyamoto AT, Tanaka S, Matloff JM. Right ventricular function during left heart bypass. J Thorac Cardiovasc Surg
1983;85:49 –53.
34. Waldenberger F, Kim YI, Laycock S, Meyns B, Flameng W.
Effects of failure of the right side of the heart and increased
pulmonary resistance on mechanical circulatory support
with use of the miniaturized HIA-VAD displacement pump
system. J Thorac Cardiovasc Surg 1996;112:484 –93.
35. Song X, Throckmorton AL, Untaroiu A, et al. Axial flow
blood pumps. ASAIO J 2003;49:355– 64.
36. Saito S, Westaby S, Pigott D, et al. Reliable long-term
nonpulsatile circulatory support without anticoagulation.
Eur J Cardiothorac Surg 2001;19:678 – 83.
37. Valdes F, Takatani S, Jacobs BG, et al. Comparison of
haemodynamic changes in a chronic non-pulsatile biventricular bypass and total artificial heart. Trans ASAIO 1980;
26:455.
38. Tatsumi E, Toda K, Taenaka Y, et al. Acute phase responses
of vasoactive hormones to nonpulsatile systemic circulation.
ASAIO J 1995;41:460 –5.
39. Golding LR, Jacobs G, Murakami T, et al. Chronic nonpulsatile blood flow in an alive awake animal: 34 days survival.
Trans ASAIO 1980;26:251.
40. Saito S, Westaby S, Pigott D, et al. End-organ function
during chronic non-pulsatile circulation. Ann Thorac Surg
2002;74:1080 –5.
41. Nojiri C, Kijima T, Maekawa J, et al. Terumo implantable left
ventricular assist system: results of long-term animal study.
ASAIO J 2000;46:117–22.
DISCUSSION
DR EVGENIJ POTAPOV (Berlin, Germany): This is a very
important study because there is a fairy tale about an increased
incidence of right heart failure in patients with axial flow pumps.
I have some questions.
You showed that there are no significant differences in right
heart dysfunction between the two groups. You showed also that
there are no significant differences in need of right ventricular
assist devices. And you reported that in the first period of your
study you used mostly pulsatile pumps and in the second period
of your study you used mostly axial flow pumps. My question is,
do you think that this fact and the improvement in the selection
of patients over time may contribute to a decrease or to the trend
toward a decrease of right heart failure in your study?
MR PATEL: Thank you very much for that important question.
We agree with what you are saying. We believe that improve-
ment in patient selection has lowered our incidence of right
heart failure.
DR ROBERT F. KORMOS (Pittsburgh, PA): Nice work, very
nicely presented, and I think it emphasizes the progression of
understanding of managing the right ventricle very nicely across
the board, and I think your comments about patient selection
probably are relevant. My question relates to the methodology
that you used to try and prevent right heart dysfunction, which
essentially is to not try to run the device at the highest possible
output that you can achieve but prevent septal shift. I agree that
is an important step. But do you think this is achievable with the
pulsatile devices, because my sense is that we are kind of doing
that already with the pulsatile systems; we are doing the same
strategy. So is it the strategy or the type of device that will impact
on prevention of right ventricular dysfunction?
ADULT CARDIAC
Ann Thorac Surg
2008;86:832– 40
840
ADULT CARDIAC
PATEL ET AL
HEARTMATE II VERSUS I LVAD
Ann Thorac Surg
2008;86:832– 40
MR PATEL: That is a great question, probably better answered
by Dr Conte.
that we cause a lot of our own problems by overflowing the right
ventricle early on.
DR CONTE (Baltimore, MD): Bob, I think you and I have dealt
with this for a while and I don’t think we know the right answer
to that. Certainly the only way we can deal with the pulsatile
flow pumps is to turn them down into a fixed rate mode, and
certainly we have done that. I think one of the problems that we
got into with the pulsatile flow pumps that led to some of the
right ventricular dysfunction problems was the fact that we let
the left ventricle become unloaded and we flew at flows much
higher than the right ventricle could stand early on, and we
caused a lot of the right ventricular dysfunction, because I don’t
think we appreciated the fact that we were overcirculating the
right ventricle early on. And as we saw, if we can get those
people through that they did fine long term, and I don’t really
think the right ventricle has proved to be a nonissue long term
once we get them through that first 2 or 3 days. But I do think
DR JOSEPH C. CLEVELAND (Denver, CO): Again, I congratulate Mr Patel and Dr Conte on a nice series of patients. Quickly,
realizing these are historical controls, one other thing to look at
I think, based upon our perspective of implanting the HeartMate
II, is the bleeding less? Did you control for blood use in the two
groups to see if there was a difference, because I think magnitude of perioperative bleeding is a contributor to right heart
dysfunction. We seem to give less blood with the HeartMate II,
and I think that may be something to consider. Is that something
you looked at as well?
MR PATEL: No, that is not something that we included in our
analysis. However, we are currently collecting those data and
agree that including the use of blood products is important in
analyzing right heart dysfunction after LVAD implantation.
Southern Thoracic Surgical Association: Fifty-Fifth
Annual Meeting
The Fifty-Fifth Annual Meeting of the Southern Thoracic
Surgical Association (STSA) will be held November 5– 8,
2008, in Austin, Texas.
Manuscripts accepted for the Resident Competition must be submitted to the STSA headquarters
© 2008 by The Society of Thoracic Surgeons
Published by Elsevier Inc
office no later than September 29, 2008. The Resident
Award will be based on abstract, presentation, and
manuscript.
Please visit www.stsa.org for more information on the
meeting and membership.
Ann Thorac Surg 2008;86:840
•
0003-4975/08/$34.00