Blood Pressure and Adverse Events During Continuous

Blood Pressure and Adverse Events During Continuous Flow Left Ventricular
Assist Device Support
Saeed et al: Blood Pressure and Adverse Events
Omar Saeed, MD1; Rita Jermyn, MD1; Faraj Kargoli, MD, MPH1;
Shivank Madan, MD1; Santhosh Mannem, MD1; Sampath Gunda, MD1;
Cecilia Nucci, RN1; Sarah Farooqui, MD3; Syed Hassan, MD3; Allison Mclarty, MD4;
Downloaded from http://circheartfailure.ahajournals.org/ by guest on November 18, 2016
Michelle Bloom, MD3; Ronald Zolty, MD1; Julia Shin, MD1; David D’Alessandro, MD2;
Daniel Goldstein, MD2; Snehal Patel MD1
1. Department of Medicine, Division of Cardiology, Montefiore Medical Center
Center,
err, Al
A
Albert
bert
be
rt
Einstein College of Medicine, Bronx, NY
2. De
Department
epa
part
rtme
rt
meent ooff Cardiothoracic Surgery, Montef
Montefiore
effio
iore Medical Center,, A
Albert
lbert Einstein College
of M
Medicine,
ed
dic
iciine,
in Bronx,
Bronx, NY
3.. D
Department
epartmeent ooff Me
Medi
Medicine,
dici
di
cine
ci
ne, Di
ne
Divi
Division
visi
vi
sion
si
on ooff Ca
Card
Cardiology,
rdiolo
rd
ogyy, Stony
Ston
St
onyy Brook
on
B oo
Br
ook University,
U iv
Un
iver
ersity
er
y, Lo
Long
ng Isl
Island,
slan
sl
and,
an
d, N
NY
Y
4 De
4.
D
Department
partment of
of Surgery,
Surrgeery, Division
Divisio
iv ionn off Cardiothoracic
Carddioothorracicc Surgery,
Sur
urggerry, Stony
Stony Brook
Sto
Broook
ook University,
Univeerssity
si y, Long
Lonng
IIsland,
Is
land, NY
Y
Correspondence to
Snehal R. Patel,, MD
Montefiore Medical Center, Albert Einstein College of Medicine
3400 Bainbridge Avenue, MAP Building 7th floor
Bronx, N.Y. 10467
Phone: 718 920-2248
Fax: 718 652-1833
Email: [email protected]
DOI: 10.1161/CIRCHEARTFAILURE.114.002000
Journal Subject Codes: Heart failure: [110] Congestive
Abstract
Background—Adverse events such as intracranial hemorrhage (ICH), thromboembolism (TE)
and progressive aortic insufficiency (pAI) create substantial morbidity and mortality during
Continuous Flow Left Ventricular Assist Device (CF) LVAD support yet their relation to blood
pressure control is underexplored.
Methods and Results—A multicenter retrospective review of patients supported for at least 30
days and up to 18 months by a CF LVAD from June 2006 to December 2013 was conducted.
All outpatient Doppler Blood Pressure (DOPBP) recordings were averaged up to the time of
ICH, TE or pAI. DOPBP was analyzed as a categorical variable grouped as high (>90 mmHg,
Downloaded from http://circheartfailure.ahajournals.org/ by guest on November 18, 2016
n=40), intermediate (80-90 mmHg, n=52) and controlled (<80 mmHg, n=31). Cumulative
survival free from an AE was calculated using Kaplan–Meier curves and Cox hazard ratios
g DOPBP ggroup
p had worse baseline renal function,, lower
(HRs)) were derived. Patients in the high
ACE inhibitor or ARB usage during CF LVAD support and a more prevalent hist
tor ooff
tory
history
hypertension. Twelve (30%) patients inn the high DOPBP group had an AE, in comparison to 7
13%
3%)) patients
3%
pati
tien
ti
entts inn tthe
en
he intermediate DOPBP group an
and only 1 (3%)) in
i thee ccontrolled
ontrolled DOPBP
(13%)
grou
grou
up. The likelihood
like
kellihood
ke
od of
of an AE
AE increased
in
incr
n rea
ease
sed inn patients
se
patieentts with
wit
ithh a high
high
ghh DOPBP
DOP
OPB
BP ((adjusted
ad
dju
just
sted
ed H
Rs [[95%
95%
95
%
group.
HRs
CI
47.3], p=
=0.012
12 vs. con
o trolledd aand
on
nd 22.6
.6
6 [[0.93-7.4],
0.93
0.
93-7
93
7.4
4], pp=0.068
=0.068
8 vvs.
s. in
nt meediat
nterme
di te)).
CI]:: 16.4 [1.8-14
[1.8-147.3],
p=0.012
controlled
intermediate).
Overal
alll, a similar
al
siimil
ilar association
il
associiati
t on was
was notedd for
for
or the
the risk
riiskk off ICH
ICH (p=0.015)
(p=
(p
=0.0015
15)) and
and
d pAI
AI (p=0.078)
(p=
p=00.07
78) but
butt not
nott
Overall,
TE (p=0.638). Patients with an AE had
had a higher
hig
ighherr DO
DOPB
PBP
PB
P (9
(90±
0±10
0±
10 mmHg) in comparison to those
DOPBP
(90±10
with
ith
thoutt an AE
AE (85±10
(85±
(8
5±10
5±
10 mmHg;
mmHg;
H p=0.05).
p=0
0.05)
05)
5).
without
Conclusions—In a population at risk, higher DOPBP during CF LVAD support was
significantly associated with a composite of adverse events.
Key Words: blood pressure, intracerebral hemorrhage, thrombus, aortic regurgitation, left
ventricular assist device
Although heart transplantation (HT) remains the gold standard treatment for end stage heart
failure, donor shortage has necessitated the development of an alternative cardiac replacement
therapy. To address this demand, the field of mechanical circulatory support has grown
tremendously over the past 15 years. Due to exorbitant wait times and stringent eligibility
criteria for HT, the number of people, both as bridge to transplantation (BTT) and destination
therapy (DT), living for extended periods on LVAD support is on the rise. Current best
estimates place the total number worldwide living with a LVAD at >7,000. As this figure is
Downloaded from http://circheartfailure.ahajournals.org/ by guest on November 18, 2016
expected to increase exponentially in the coming years, refinement in long term management
strategies is needed.
The newest generations of LVADs are rotary blood pumps utilizing either
eitheer ce
cent
ntri
nt
rifu
ri
fuga
fu
gall or
ga
centrifugal
axial pr
ppropulsion.
op
pul
u si
sion.. T
hese designs offer significant ad
adva
v ntages including ppump
u p miniaturization,
um
These
advantages
ilennt operation,
operatiion
on,, and
and most
most importantly,
imp
mpor
orta
or
tant
ta
ntly
nt
ly
y, enhanced
enha
en
hanc
ha
nced ddevice
nc
ev
vicee du
dura
rabi
ra
b liity.
ty 1 Si
Since
inc
ncee thes
these
esee pu
es
pump
pumps
mpss mo
mp
move
ve
silent
durability.
blood
blloood from the left
lef
eftt ventricle
ef
venntr
ntriclee too the aorta
aor
o ta throughout
throougghou
ut thee cardiac
ca diaac cycle,
card
cy
ycle, th
they
heyy create
crreaate a nnono nonphysiologic continuous blood flow
and
w pa
ppattern
ttern an
nd ar
aree referred
r fe
re
ferr
rred
rr
e to as continuous flow (CF) CF
LVADs.
LVAD
LV
ADs.
s. With
With reduced
red
educ
uced
ed or
or absent
abse
ab
sent
ntt arterial
art
rter
eria
iall pu
ppulse
lsee pr
ls
ppressure,
essu
es
sure
re,, tr
trad
traditional
adit
itio
iona
nall me
meth
methods
thod
odss of nnon-invasive
on-iinv
on
nvas
asiv
ivee
blood pressure measurements are unreliable. This limitation is overcome by utilizing Doppler
ultrasound of the brachial artery after an arm cuff is deflated and recording the audible
restoration of flow as the Doppler blood pressure (DOPBP).2 DOPBP has previously been
shown to have excellent correlation to systolic blood pressure measurement via arterial line in
CF LVAD subjects.3
Of the most common significant adverse events associated with CF LVADs, 3 are
potentially modulated by blood pressure. These are (1) intracranial hemorrhage, (2)
thromboembolic events, and (3) development of aortic insufficiency. The recently published
ISHLT guidelines for Mechanical Circulatory Support acknowledge this potential by
recommending that patients with nonpulsatile CF LVADs should have DOPBP goal of ”
80mmHg.4 This recommendation is based on a Level of Evidence C (expert opinion),
highlighting the need for an evidence based rationale of optimal blood pressure targets in this
population. Therefore, in the current study, we aimed to examine the association of DOPBP and
adverse events in patients supported with a CF LVAD.
Downloaded from http://circheartfailure.ahajournals.org/ by guest on November 18, 2016
Methods
Study Population
We retrospectively reviewed all patients undergoing CF LVAD placement from Ju
June
n 11,, 20
ne
2006
06 to
December 31, 2013
201
013 at Montefiore Medical Center and
an Stony Brook Univ
University
versity Medical Center.
Patients
Pati
Pa
ients weree in
included
ncl
clud
u ed iiff th
they
e ssurvived
ey
urvi
ur
vive
vi
vedd at
ve
at lea
least
east
ea
stt 30 da
days oon
n CF L
LVAD
VAD
VA
D su
support
upp
pport an
andd we
w
were
ree
discharged
di
isccha
h rged to thee outpatient
outp
pattient setting.
set
etting. Ass pper
er ins
iinstitutional
nstituutio
ionaal pprotocols,
rootocolss, mode
moderate
eratee orr grea
greater
ater aaortic
ortic
insufficiency
nsufficiency was surgically addre
addressed
esssed
e at the ti
ttime
me ooff LV
LVAD imp
implant
plant and these patient were
excluded. The studyy protocol
protocol was approved
ap
ppr
p oved byy the institutional review boards of both centers.
Data collection
Pre-operative clinical information and baseline demographics were collected from medical
charts. The analysis commenced at 30 days post implant and after hospital discharge. DOPBP
recordings from all outpatient clinical visits were averaged up to the time of the following
adverse events (AEs): 1) intracranial hemorrhage (ICH), 2) thromboembolic events (TEs), and 3)
development of moderate or severe aortic insufficiency, hence forth referred to as progressive
aortic insufficiency (pAI). If patients did not have an ICH, TE or pAI then DOPBP was
averaged until transplantation, device explantation, expiration or up to 18 months on CF LVAD
support. INR was retrieved at the time of ICH and TE.
Definition of Adverse Events
Intracranial hemorrhage was defined as the presence of hemorrhage noted on a head CT scan
accompanied by neurological symptoms. Thromboembolic events included ischemic CVA,
peripheral embolism or device thrombosis. Ischemic CVA was defined as the presence of acute
Downloaded from http://circheartfailure.ahajournals.org/ by guest on November 18, 2016
cerebral infarction noted on a head CT scan accompanied by neurologic symptoms. Peripheral
embolism was confirmed by radiological imaging. Device thrombosis was defined as the
presence of thrombus within the LVAD leading to device malfunction and explan
explantation.
ntat
a io
on. Al
Alll
device thromboses
thrombose
ses were subsequently confirmed byy direct inspection andd visualization of the
pump.
pum
pu
mp. Development
Devel
elloppme
mentt of
of pAI
pAI was
was determined
dete
de
term
te
rmin
rm
i ed by
by itss presence
pressen
ence
ce on
on clinical
clin
cl
in
nic
icall echocardiograms
ech
c ocar
ardi
ar
diog
di
ograms
og
ms (based
(ba
base
sed
se
onn tthe
h Americann S
he
Society
ocieety of E
Echocardiography
c ocarrdi
ch
d og
ographhy crit
criteria
terria ffor
or m
moderate
oderaate or sev
severe
everee A
ev
AI)
I) pe
performed
erfforrmed
during the course of CF LVAD support.
sup
uppo
up
p rt.
po
Data analysis
Patients were categorized into three groups based on their average outpatient DOPBP: 1) high
(•90 mmHg), 2) intermediate (80-89 mmHg) and 3) controlled (<80 mmHg). The primary
endpoint was the occurrence of any AEs. Secondary endpoints were the occurrence of each
individual AE, including ICH, TEs, or pAI. If patients had more than one AE, then analysis of
the primary endpoint was done with the time to the first AE and the time to each individual AE
was used in the secondary endpoint analysis.
Statistical analysis
Data are displayed as means ± standard deviation. Baseline demographics were compared
between patients in all three DOPBP groups by Analysis of Variance (ANOVA) for continuous
variables and the chi-square test for categorical variables. Cumulative survival free from
primary and secondary endpoints was shown using Kaplan–Meier curves and hazard ratios were
calculated by univariable and multivariable Cox proportional analysis. Multivariable analysis
was adjusted for variables that may have clinically affected the outcome of interest including
Downloaded from http://circheartfailure.ahajournals.org/ by guest on November 18, 2016
age, hypertension, diabetes mellitus, atrial fibrillation, aspirin use, baseline renal function,
gastrointestinal bleeding on CF LVAD support, and device type. P values of ”0.05 were
considered statistically significant. DOPBP was also compared as a continuous vari
vvariable
ari
riab
ab
ble bbetween
etw
et
ween
patients with andd without the combined endpoint byy the Student’s t-test. Statistical
Statistical analyses
were
w
ere conducted
conductted
d in
in SAS
SA
AS version
vers
ve
r ionn 9.3
rs
9.3 (SAS
(SAS
A Institute,
Ins
nsti
ns
titutee, C
Cary,
ary
ry
y, NC
NC).
)
Results
R
Re
sult
su
ls
lt
Patient Characteristics
The study group was comprised of 123 patients who were supported by a CF LVAD for at least
30 days and met the inclusion criteria. Nearly half (49%) of the total cohort had an ischemic
etiology of cardiomyopathy and 31% had atrial fibrillation. Table 1 demonstrates the clinical
characteristics of these patients categorized into the three DOPBP groups. Patients in the greater
DOPBP groups had a higher baseline creatinine (p<0.003) and were more likely to have a history
of hypertension (p<0.001). Non-white patients comprised the majority in the high DOPBP group
(64%) in contrast to the controlled DOPBP group which consisted mainly of Whites (62%).
Aspirin and beta blocker use was high and similar across the groups. ACE inhibitor and ARB
utilization decreased with each successive DOPBP group while use of the alternative
vasodilators hydralazine/nitrate tended to increase (Table 1), perhaps corresponding to
differences in the baseline renal function.
On average, there were 10 ± 7 DOPBP readings per patient in the entire cohort during the
study period without a significant difference between groups (Table 1). In those patients who
had an AE (n=20), the mean number of readings was 6 ± 4. It is important to note that the
readings at the time of the event were excluded since BP is known to be artificially elevated
Downloaded from http://circheartfailure.ahajournals.org/ by guest on November 18, 2016
especially acutely during a neurological episode.
Prevalence of Adverse Events (AEs)
Twenty
y (16%)
(16%) patients
pat
atients in the entire group had a predefined
pred
edefined adverse event
even
nt during the follow up
period
peri
pe
iod (109 pa
ppatient
tien
ti
ent ye
en
year
years
arss of LV
ar
LVAD
VAD ssupport).
uppo
up
p rtt). Ei
po
E
Eight
gh
ht (6.5
(6.5%)
5%) ppatients
a ieents
at
nts ex
eexperienced
peri
pe
rienceed an IICH,
ri
CH
H, 7
(5.7%)
5.77%)
% had a TE
E aand
nd 6 ((4.9%)
4.9%
%) demon
%)
demonstrated
onsttraated pA
on
pAI.
AI. T
The
hee me
m
median
diian
n tim
time
me to fi
first
i rst A
irs
AE
E was 922 (ra
(range:
range::
ra
31-513) days DOPBP. Figure 1 displays
disp
di
splays
sp
y the distribution
disttri
ribu
b ti
tion
on of each type of AE amongst the three
DOPBP groups.
g oups
gr
p . Onlyy 1 of 31 (3.2%)
( .2%)) patients
(3
patients in the controlled DOPBP group
g oupp had an AE – an
gr
ischemic CVA. The prevalence of AEs increased within higher DOPBP groups: there were 7
(13.4%) events in the intermediate group and 12 (30%) events in the high group. The average
INR two weeks prior to ICH was 2.2 (range 1.2-5.8; 1 patient was above the institutional goal of
2-3) and two weeks prior to TE it was 2.4 (range 1.4-4.8; 3 patients were <2). There were 12
patients included with a centrifugal device and only 1 (8%) had an AE (ICH). In comparison, 19
out of the 111 (17%) patient with an axial flow device had AEs (p=0.69).
Risk of Intracranial Hemorrhage, Thromboembolic Events and Progressive AI
At the end of follow up, there was a significant difference in the occurrence of the primary
endpoint between DOPBP groups. Survival free from ICH, TE or pAI at 18 months after CF
LVAD implantation was 70% in the high DOPBP group, 86% in the intermediate and 97% in the
low DOPBP groups (p=0.004; Figure 2). Patients with a high DOPBP had an adjusted hazard
ratio (aHR) of 16.4 [95% CI: 1.8-147.3] (vs. controlled DOPBP) and of 2.6 [95% CI: 0.93-7.4]
(vs. intermediate DOPBP) for the combined endpoint. When DOPBP was assessed as a
Downloaded from http://circheartfailure.ahajournals.org/ by guest on November 18, 2016
continuous variable, patients with a combined endpoint had a significantly higher DOPBP of
90±10 in comparison to those without an event 85±10 (p = 0.05).
Both univariable HRs (Table 2) and survival analysis (Figure 2) showed a graded
grade
dedd rise
rise in
in
the
he risk
sk of
of composite
co
omp
pos
osite AEs. In individual endpoint analysis
ana
n lysis by Kaplan Meier
Meie
Me
i r curves, survival free
from
fr
rom
m ICH at 18 m
months
o th
on
thss wa
wass sign
significantly
gnif
gn
ific
if
ican
ic
ntl
tlyy lo
lower
owe
wer in the
thhe high
hig
ighh DOPBP
D PB
DO
PBP
P group
g ou
gr
oupp (85%)
(85%
%) in comparison
com
mpa
pari
riso
ri
sonn
so
to
o in
intermediate
ntermediate (96
((96%)
96%) an
and
nd lo
low
ow (100%)
% D
DOPBP
OPBP
BP gr
groups
roupss (p
(p=0
(p=0.015;
=0
0.0
015;; F
Figure
iguree 2).
2). There
Th was
wa noo
significant
ignificant difference in survival free
fre
reee from TEs
TE
Es across
acro
ac
r sss D
DOPBP
OPBP groups (p=0.638; Figure 2).
Proportionally
Prop
portionally
y fewer ppatients
atients were free of ppAI
AI att 18 months in the high
g DOPBP gr
ggroup
oupp (9
((90%)
0%)) in
comparison to intermediate (96%) and low (100%) DOPBP groups (p=0.078; Figure 2).
Clinical Outcomes
Patient’s clinical outcomes with and without adverse events are listed in Table 3. Six out of
eight patients (75%) expired after an ICH and the remaining two stayed on CF LVAD support.
Of the three patients with an ischemic CVA, 1 expired, 1 underwent cardiac transplantation and
the remaining patient was kept on CF LVAD support. Two of the four patients with pump
thrombosis underwent emergent device exchange while the other two patients had cardiac
transplantation. No patients with progressive AI had any aortic valve interventions. Importantly,
103 (84%) patients did not have an ICH, TE or pAI and 43 of them underwent cardiac
transplantation, 47 remained on CF LVAD support, 3 underwent device exchange and 10
eventually expired.
Discussion
The results of this study demonstrated that there is a graded association of blood pressure on a
Downloaded from http://circheartfailure.ahajournals.org/ by guest on November 18, 2016
composite of adverse events including intracranial hemorrhage, thromboembolic events and
progressive aortic insufficiency in patients supported by a CF LVAD. This was a contemporary
cohort with a mean age of 57, 49% ischemic etiology, 24% female with diligent ffollow
ollo
ol
low
lo
w up iin
n our
LVAD clinic as rreflected
e lected by their aggressive medica
ef
medical
cal management: almo
almost
ost
s 90% were on a beta
blocker,
bloc
bl
cker, moree th
than
han ½ oon
n AC
ACE in
inhi
inhibitor/ARB,
hibi
hi
bito
bi
or/
r AR
ARB,
B and aapproximately
B,
ppro
oxi
xima
mately
ma
y 1/3
/3 oon
n a comb
combination
mbin
mb
inat
in
atio
at
i n of
hydralazine/nitrates.
hy
ydr
dralazine/nitraatees. IIn
n the an
anal
analysis,
a ysis,, w
wee dem
demonstrated
monstrrat
ated tthat
hat pa
ha
patien
patients
ntss withh a me
mean
an DOP
DOPBP,
OPBP
OP
BP, with
BP
within
hin
n
the
he guideline recommendation off <8
<80mmHg,
g, wer
were
eree rema
er
remarkably
markably
ma
y free of AEs with only 1 of 31
subjects
ubjjects experiencing
expperiencingg an event. The risk of an AE increased byy DOPBP: the intermediate group
group
p
(80-89mmHg) had an aHR of 2.6 (p=0.06) which rose to 16.1 (p<0.012) in the high group
(>90mmHg) in comparison to the controlled group. Secondary outcomes analysis of the
individual endpoints demonstrated that this association was maintained for ICH and pAI. Such
findings must be interpreted with an understanding that patients with a high DOPBP comprised a
population with greater risk factors for AEs. Patients in the high DOPBP group had worse
baseline renal function, lower usage of ACE inhibitors or ARBs, and more had a history of
hypertension in comparison to those in the intermediate and controlled groups.
Notwithstanding such intrinsic risk factors, this is the first report to associate DOPBP
with a composite of adverse events in a CF LVAD population.
The significant design advantages offered by rotary pumps suggest that not only current
but also all future generations of ventricular assist devices will utilize this mechanism for blood
propulsion. These advantages present new challenges as well, a notable one being the creation
of a distinct non-pulsatile circulatory physiology. In patients with a CF LVAD, measurement,
impact on outcomes, and target ranges for treatment of blood pressure currently lack clarity.
Downloaded from http://circheartfailure.ahajournals.org/ by guest on November 18, 2016
Therefore, in the current analysis, we sought to define DOPBP goals in CF LVAD subjects by
exploring the potential impact of blood pressure on adverse events. In particular, blood pressure
could theoretically have a unique contribution to the development of ICH, thrombo
thromboembolic
boem
bo
embo
em
boli
bo
licc
li
events,, and aortic
ic insufficiency. Although these hypotheses
hy
ypo
p theses may appear ev
eevident,
ident, they have never
previously
prev
pr
viously bee
been
e n va
valida
validated.
date
da
ted.
te
d.
In order tto
o mech
mechanistically
chanisttica
ch
tica
cally un
unde
understand
erstannd
nd the as
association
ssooci
ciaatio
on bet
between
tween blo
bblood
loodd pressure
pressuure and
pr
nd
adverse events demonstrated in the
he ppresent
resent ana
analysis,
n lys
ysis
ys
is, an examination
is
examination of each individual endpoint
iss relevant. There is an established causal relationshipp between elevated blood pressure
pressure and ICH
in normal pulsatile physiology.5 We hypothesized that the effect of a blood pressure elevated
continuously throughout the cardiac cycle on the intra-cerebral vasculature may be even more
detrimental. In our cohort, no patients with a controlled DOPBP experienced an ICH whereas 6
of 40 (15%) patients with a DOPBP • 90mmHg had an event during the follow-up period. ICH
was a devastating complication with 6 of 8 patients expiring. Furthermore, despite excluding
events within the first 30 days after implant, ICH tended to occur early with a median time to
event in the entire cohort of 69 (range:31-226) days. These findings highlight the importance of
early blood pressure control to prevent this serious complication.
Development of aortic insufficiency during CF LVAD support is common with freedom
from AI at one year documented as approximately 75%.6,7 Aortic valve closure appears to play a
primary role in this process and while the pathophysiology remains unclear it is likely as a result
of an increase in the duration of the instantaneous transvalvular pressure gradient leading to
pathological remodeling and commissural fusion.8,9 Aortic root dilatation as a consequence of
the elevated pressure caused by retrograde flow from the outflow cannula is also a contributing
factor.10 Appropriately, much attention has been dedicated to pump speed operation for
Downloaded from http://circheartfailure.ahajournals.org/ by guest on November 18, 2016
promotion of aortic valve opening to mitigate this complication. Less attention has focused on
hypertension which could potentially influence the aforementioned pathophysiology via
increased
ncreased aortic pressure and contribute to pAI. We found a strong trend towardss in
incr
increased
crrea
ease
sedd pA
se
ppAII
with higher
hig
gher DOPBP:
DOPB
PBP: none of the patients in the cont
controlled
ntrolled DOPBP groupp while 4% in the
intermediate
nteermediate gro
group
roup
ro
up and
nd 110%
0% inn th
thee hi
hhigh
gh ggroup
ro
oup dev
developed
velopeed mo
mode
moderate
d ra
rate
te tto
o se
seve
severe
vere A
ve
AII (p
(p=0
(p=0.078)
=0.0
=0
078
78))
during
du
uriing
n follow up
up.
p. Importantly,
Imp
portanttly
ly,, the pr
pprogression
og
gressiion
n occ
occurred
curreed llate
atee aatt a m
median
edian
n ooff 3347
47 (range:
47
(rang
nge:
ng
e 10
102-513)
02-5
513
days emphasizing the importancee of
of pe
ppersistent
rsistent
nt blood
blo
lood
lo
o pressure
pre
ressure control during extended support.
In the secondaryy endpoint
endpo
p int analysis,
analys
y is,, a greater
g eater proportion
gr
p opportion of patients
pr
p tients had TE events in the
pa
high (7.5%) and intermediate (5.8%) compared with the controlled group (3.2%) but this did not
reach statistical significance. We had hypothesized that higher blood pressure by increasing
afterload on the pump would reduce pump flow and increase propensity for device thrombus that
could manifest as either embolic events or pump malfunction. A similar hypothesis had
previously been validated in an analysis of the HeartWare ADVANCE bridge to transplant trial
where elevated blood pressure increased the risk of pump thrombus.11 Our discordant findings
from the HW analysis may be explained by the high proportion of the axial flow HMII (90%) in
our study. Centrifugal pumps, such as the HW, are known to be more afterload sensitive and
therefore will have a greater reduction in pump flow in response to hypertension.12 The variable
clinical impact of blood pressure on the HMII and HW raise important questions that need to be
answered in a larger population.
There are several limitations to the current study that are worth addressing. First, patients
with a high DOPBP also had demographic characteristics which may have put them at a higher
risk of having an AE, thereby mitigating some of the risk attributed to DOPBP control. Second,
we chose to include both of the clinically approved devices, HMII and HW, in our analysis. The
Downloaded from http://circheartfailure.ahajournals.org/ by guest on November 18, 2016
afterload sensitivity of these pumps is different based upon their axial versus centrifugal design
and therefore, as previously noted, the effect of hypertension on the devices maybe variable.
This is a retrospective study and we were limited to blood pressure readings obta
obtained
ain
ined
ed during
dur
urin
ingg
in
clinic visits, although
altho
hough it should be noted that we maintain
mai
aintain close follow up
p with our patients who
are ggenerally
ar
enerallyy se
seen
een onn a bi
biwe
biweekly
week
we
kly tto
o mo
mont
monthly
nthlly ba
nt
basis.. A
Also,
lsoo, as iiss co
common
omm
mmon
onn tto
o many
ny L
LVAD
V D an
VA
anal
analyses,
alys
al
yses
ys
the
he in
iinfluence
fluence of ccenter
enterr sspecific
pecifficc manag
management
a em
ement st
strategies
trate
a egiies m
may
ay lim
limit
mit tthe
he gen
generalizability
nerraliiza
izabilityy ooff tthese
hese
findings. INR was not retrievablee inn patients
patientss without
w th
wi
thou
o t AEs
AE and thus it is not included in the
multivariable analys
analysis.
y is. However,, the degr
degree
g ee off anticoagu
anticoagulation
g lation is not likely
y to impact
impa
p ct AEs
because the average INR prior to AEs was within the therapeutic range. Patients with the highest
DOPBP had worse renal function at baseline, which may have limited certain anti-hypertension
therapies and is itself a marker of a worse prognosis. Finally, this was a small cohort with a low
event rate, thereby limiting the utility of multivariable modeling and requires confirmation in a
larger population.
In conclusion, our data demonstrates that in a population at risk, there is a graded
association between poor blood pressure control and adverse events. As the field of mechanical
circulatory support moves to longer durations of support this issue will become increasingly
important. Further study in a larger population exploring this association as well as the
difference between centrifugal and axial flow design and the efficacy of antihypertensive
treatment is warranted.
Sources of Funding
This study was supported by intramural research funds.
Downloaded from http://circheartfailure.ahajournals.org/ by guest on November 18, 2016
Disclosures
Dr. Goldstein serves as a consultant for Thoratec Inc. He also serves on the scientific advisory
board of HeartWare and is their surgical proctor.
Referenc
References
nces
nc
1.
1.
2.
3.
4.
5.
Slaughterr M
MS,
S, Roger
R
Rogers
o rs JJG,
G, Mila
Milano
anoo CA
CA,
A, Rus
Russell
sseell SD
SD, Co
Contee JJV,
V, Fe
Feldman
eld
dmann D, S
Sun
un
nB
B,,
Tatooles
Tato
to
ool
olees A
AJ,
J, D
Delgado
ellgadoo R
RM
M 3r
3rd,
d L
Long
ong JW
ong
JW,
W, Wo
Wozn
Wozniak
zn
niaak TC
TC,, G
Ghumman
hum
ummaan W, F
um
Farrar
arraar DJ
DJ, F
Frazier
razzieer
OH;.
OH Advanced
Ad
d heart
heart failure
fai
ailu
lure treated
treate
tedd with
te
wiith continuous-flow
conti
t nuuous-fl
ti
flow left
fl
lef
eft ventricular
venttric
i ullar aassist
ssistt ddevice.
evice. N
Engl J Med. 2009; 361:22
361:2241-2251.
241
41-2
-225
-2
251.
25
1
D,, S
Shah
A,, R
Russell
Bennett MK, Roberts CA, Dordunoo
Dor
ordu
duno
du
nooo D
no
hahh A
ha
usse
us
sell
se
ll SD. Ideal methodology to assess
systemic
ventricular
assist
systtemiic blood
bloodd pressure in
in patients
pattientts with
wiith continuous-flow
conttinuous-fflow lleft
eft vent
tricullar as
siistt ddevices.
eviices. J
Heart Lung Transplant. 2010; 29:593-594.
Lanier GM, Orlanes K, Hayashi Y, Murphy J, Flannery M, Te-Frey R, Uriel N,
Yuzefpolskaya M, Mancini D, Naka Y, Takayama H, Jorde UP, Demmer RT, Colombo
PC. Validity and Reliability of a Novel Slow Cuff-Deflation System for Noninvasive
Blood Pressure Monitoring in Patients with Continuous-Flow Left Ventricular Assist
Device. Circ Heart Failure. 2013; 6:1005-1012.
Feldman D, Pamboukian SV, Teuteberg JJ, Birks E, Lietz K, Moore SA, Morgan JA,
Arabia F, Bauman ME, Buchholz HW, Deng M, Dickstein ML, El-Banayosy A, Elliot T,
Goldstein DJ, Grady KL, Jones K, Hryniewicz K, John R, Kaan A, Kusne S, Loebe M,
Massicotte MP, Moazami N, Mohacsi P, Mooney M, Nelson T, Pagani F, Perry W,
Potapov EV, Eduardo Rame J, Russell SD, Sorensen EN, Sun B, Strueber M, Mangi AA,
Petty MG, Rogers J. The 2013 International Society for Heart and Lung Transplantation
Guidelines for mechanical circulatory support: executive summary. J Heart Lung
Transplant. 2013; 32:157-187.
Brott T, Thalinger K, Hertzberg V. Hypertension as a risk factor for spontaneous
intracerebral hemorrhage. Stroke. 1986; 17:1078-1083.
6.
7.
8.
9.
Downloaded from http://circheartfailure.ahajournals.org/ by guest on November 18, 2016
10.
11.
12
12.
2.
Pak SW, Uriel N, Takayama H, Cappleman S, Song R, Colombo PC, Charles S, Mancini
D, Gillam L, Naka Y, Jorde UP. Prevalence of de novo aortic insufficiency during longterm support with left ventricular assist devices. J Heart Lung Transplant. 2010;
29:1172-1176.
Cowger J, Pagani FD, Haft JW, Romano MA, Aaronson KD, Kolias TJ. The
development of aortic insufficiency in left ventricular assist device-supported patients.
Circ Heart Fail. 2010; 3:668-674.
Samuels LE, Thomas MP, Holmes EC, Narula J, Fitzpatrick J, Wood D, Fyfe B,
Wechsler AS. Insufficiency of the native aortic valve and left ventricular assist system
inflow valve after support with an implantable left ventricular assist system: signs,
symptoms, and concerns. J Thorac Cardiovasc Surg. 2001; 122:380-381.
Balachandran K, Sucosky P, Jo H, Yoganathan AP. Elevated cyclic stretch alters matrix
remodeling in aortic valve cusps: implications for degenerative aortic valve disease. Am J
Physiol Heart Circ Physiol. 2009; 296:H756-764.
May-Newman K, Hillen B, Dembitsky W. Effect of left ventricular assist device outflow
conduit anastomosis location on flow patterns in the native aorta. ASAIO J. 2006; 52:132139.
Najjar SS, Slaughter MS, Pagani FD, Starling RC, McGee EC, Eckman P,, Tatooles
Tat
a ooole
less AJ
AJ,
Moazami N, Kormos RL, Hathaway DR, Najarian KB, Bhat G, Aaronsonn KD
KD,, Bo
Boyc
ycee
yc
Boyce
SW. An analysis of pump thrombus events in patients in the HeartWare ADVANCE
ADV
DVAN
DV
ANCE
AN
CE
bridge
g to tr
transplant and continued access pr
pprotocol
o ocol trial. J Heart Lung
ot
Lu Transplant. 2014;
33
33:23-34.
3:23:2
233-344.
Salamo
Salamonsen
m ns
nsen
e R
RF,
F, M
Mason
a on
as
nD
DG,
G, A
Ayre
yre PJ
yr
PJ. Re
R
Response
spponsee ooff ro
rotary
y bblood
lood
lo
od ppumps
u ps tto
um
o ch
change
changes
gess in
ge
afterload
speed
are
unphysiological
when
compared
with
preload and
and af
fter
terloadd at
at a ffixed
ixe
x d sp
peedd ssetting
ettin
ng ar
re un
unph
physio
ph
olo
ogicall w
hen
en com
mpa
paredd w
ith
h
natural
heart.
Artif
2011;
the natura
al hea
art. Arti
art.
if Organs.
O gans
Or
n . 20
ns
011;; 335:E47-53.
5:E47
47-53.
47
3.
Table 1. Patient Characteristics
DOPBP Groups
Controlled
(<80 mmHg)
n=31
Intermediate
(80-90 mmHg)
n=52
High
(• 90 mmHg)
n=40
Total
n= 123
P
value
Age (years)
57±14
55±13
61±10
57±13
0.186
Female n (%)
7 (23)
16 (30)
6 (15)
29 (24)
0.383
Hypertension, n (%)
7 (23)
23 (44)
28 (70)
58 (47)
<0.001
Diabetes mellitus, n (%)
9 (29)
24 (46)
22 (55)
55 (45)
0.088
Ischemic Etiology, n (%)
14 (45)
22 (42)
24 (60)
60 (49)
0.223
1.27±0.53
1.38±0.58
1.70±0.50
1.47±0.57
0.003
11 (35)
11 (21)
16 (40)
30 (31)
0.124
Clinical
Downloaded from http://circheartfailure.ahajournals.org/ by guest on November 18, 2016
Serum Creatinine (mg/dL)
Atrial Fibrillation, n (%)
0.421
Race
Hispanic, n (%)
5 (16)
14 (27)
9 (23)
28 (23)
(23
23))
African American, n (%)
6 (19)
13 (25)
13 (33)
32 (26)
(26
26))
Wh
White,
hit
ite,
e,, n ((%)
%)
19 (62)
21 (40)
15 (37)
(37
37))
55 (45)
Other, n ((%)
Ot
%)
1 (3)
4 (8)
3 (8
(8)
8)
8 (6)
6.5±1.1
6.5±
6.
±1.1
6.7±1.2
6.7±
7±
±1.
1.2
2
7±1.2
7±1
1.2
2
6.6±1.1
6.6±
6±1.
6±
11
0.961
0
.9
96
19±9
19
9±9
20±7
20
±7
21±6
21
±6
20±7
20
±7
7
0.275
0
0.
27
Aspirin,
Asp
pirin,, n (%
((%))
26 (84)
(84
84))
42 (81)
(81
81))
27 (68)
(68))
95 (77)
(77))
0.192
0.19
ACEI and/ or ARB
ARB, n (%)
18 (58)
35 (67)
15 (38)
65 (55)
0.016
0
01
Beta blocker, n (%)
27 (81)
48 (92)
33 (83)
108 (88)
0.341
Hydralazine, n (%)
8 (26)
25 (48)
19 (48)
51 (41)
0.101
Nitrates, n (%)
3 (10)
14 (27)
12 (30)
29 (24)
0.092
Number of DOPBP readings
10±8
10±7
9±6
10±7
0.482
Patients with GIB
7 (23)
14 (27)
18 (45)
39 (32)
0.082
28 (23)
47 (38)
36 (29)
111 (90)
0.927
rotor speed (rpm)
8981±414
8927±298
8950±293
8948±327
0.810
Centrifugal***, n (%)
3 (2)
5 (4)
4 (3)
12 (9)
0.875
rotor speed (rpm)
2710±70
2788±198
2834±84
2797±144
0.570
Echocardiographic
E
chocardiograp
ho
phi
hc
LV
LVEDD,
VEDD, cm
L
LVEF,
VEF,
VE
F, %
Medications*
CF LVAD type
Axial**, n (%)
*=medications were retrieved at the time of an adverse event or at the end of follow up; LVEDD=left ventricular
end diastolic diameter; LVEF=left ventricular ejection fraction; ACEI=angiotensin converting enzyme inhibitor;
ARB=angiotensin receptor blocker; GIB=gastrointestinal bleeding; **All Heart Mate II; *** All HVAD.
Table 2. Univariable hazard ratio estimates for the risk of an adverse event
Doppler blood pressure group
Univariable
hazard ratio (95% CI)
Combined Endpoint
Intermediate vs. controlled
4.213 (0.518-34.27)
High vs. intermediate
2.727 (1.072 – 6.941)*
High vs. controlled
11.455 (1.488- 88.192)*
Intracranial Hemorrhage
Intermediate vs. controlled
High vs. intermediate
Downloaded from http://circheartfailure.ahajournals.org/ by guest on November 18, 2016
High vs. controlled
**
4.357 (0.878-21.615)
**
Thromboembolic events
Intermediate vs. controlled
1.77 (0.184-17.019)
High vs. intermediate
1.607 (0.322-8.005)
High vs. controlled
2.786 (0.289- 26.849)
Progressive Aortic Insufficiency
Intermediate
In
nte
term
rmed
rm
ediiat
ed
iate vvs.
s.. controlled
High
Hi vs. intermediate
int
ntterrme
mediiat
atee
H
High
igh vs. contro
controlled
ollled
***
2.772
2.77
2.
772 (0
77
(0.5
(0.506
.506 – 115.193)
.5
5.19
93)
**
*p
*p”0.05;
p”0.
0.05
05;; **=n
05
**=no
=no
=n
o co
comparison
ompar
mp riso
on was
as ppossible
ossi
os
sibl
si
ble du
bl
duee to no
o ev
even
events
ents
en
t iin
n th
thee
control
lledd blood pressure group
controlled
Table 3. Patient outcomes with and without adverse events
Adverse Event
Intracranial Hemorrhage (n=8)
Outcome
6 expired, 2 remained on CF LVAD support
Thromboembolic Events
Pump thrombosis (n=4*)
1 expired, 1 cardiac transplantation, 1 remained on CF LVAD
support
2 device exchanges, 2 cardiac transplantations
Peripheral embolism (n=1)
Right coronary artery aspiration thrombectomy
Ischemic CVA (n=3)
Downloaded from http://circheartfailure.ahajournals.org/ by guest on November 18, 2016
Progressive Aortic insufficiency (n=6)
All 6 underwent clinical monitoring, no aortic valve interventions
No ICH, TEs or pAI (n=103)
43 cardiac transplantations, 47 remained on CF LVAD support,
3 device exchanges (2 driveline infections, 1 driveline fracture), 10
expired.
*=1 patient’s presentation for pump thrombosis was an ischemic CVA.
Figure Legends
Figure 1. Figure 1: (A) Distribution of combined and individual adverse events across blood
pressure groups. (B) Distribution of the type of thromboembolic events across blood pressure
groups.
Figure 2. Kaplan Meier curves showing the survival free from adverse events.
Downloaded from http://circheartfailure.ahajournals.org/ by guest on November 18, 2016
Downloaded from http://circheartfailure.ahajournals.org/ by guest on November 18, 2016
Downloaded from http://circheartfailure.ahajournals.org/ by guest on November 18, 2016
Downloaded from http://circheartfailure.ahajournals.org/ by guest on November 18, 2016
Blood Pressure and Adverse Events During Continuous Flow Left Ventricular Assist Device
Support
Omar Saeed, Rita Jermyn, Faraj Kargoli, Shivank Madan, Santhosh Mannem, Sampath Gunda,
Cecilia Nucci, Sarah Farooqui, Syed Hassan, Allison Mclarty, Michelle Bloom, Ronald Zolty, Julia
Shin, David D'Alessandro, Daniel Goldstein and Snehal R. Patel
Circ Heart Fail. published online April 13, 2015;
Circulation: Heart Failure is published by the American Heart Association, 7272 Greenville Avenue, Dallas, TX 75231
Copyright © 2015 American Heart Association, Inc. All rights reserved.
Print ISSN: 1941-3289. Online ISSN: 1941-3297
The online version of this article, along with updated information and services, is located on the
World Wide Web at:
http://circheartfailure.ahajournals.org/content/early/2015/04/13/CIRCHEARTFAILURE.114.002000
Permissions: Requests for permissions to reproduce figures, tables, or portions of articles originally published in
Circulation: Heart Failure can be obtained via RightsLink, a service of the Copyright Clearance Center, not the
Editorial Office. Once the online version of the published article for which permission is being requested is located,
click Request Permissions in the middle column of the Web page under Services. Further information about this
process is available in the Permissions and Rights Question and Answer document.
Reprints: Information about reprints can be found online at:
http://www.lww.com/reprints
Subscriptions: Information about subscribing to Circulation: Heart Failure is online at:
http://circheartfailure.ahajournals.org//subscriptions/