Analytic Review

Journal of Intensive Care Medicine

2017, Vol. 32(5) 312-319
Bivalirudin for Alternative Anticoagulation ª The Author(s) 2016
Reprints and permission:
sagepub.com/journalsPermissions.nav
in Extracorporeal Membrane Oxygenation: DOI: 10.1177/0885066616656333
journals.sagepub.com/home/jic
A Systematic Review

Filippo Sanfilippo, MD, PhD1,2, Sven Asmussen, MD3,4,

Dirk M. Maybauer, MD, PhD4,5, Cristina Santonocito, MD1,
John F. Fraser, MD, PhD4, Gabor Erdoes, MD6,
and Marc O. Maybauer, MD, PhD4,5,7

Abstract
Background: Extracorporeal membrane oxygenation (ECMO) offers therapeutic options in refractory respiratory and/or
cardiac failure. Systemic anticoagulation with heparin is routinely administered. However, in patients with heparin-induced
thrombocytopenia or heparin resistance, the direct thrombin inhibitor bivalirudin is a valid option and has been increasingly
used for ECMO anticoagulation. We aimed at evaluating its safety and its optimal dosing for ECMO. Methods: Systematic web-
based literature search of PubMed and EMBASE performed via National Health Service Library Evidence and manually, updated until
January 30, 2016. Results: The search revealed 8 publications relevant to the topic (5 case reports). In total, 58 patients (24
pediatrics) were reported (18 received heparin as control groups). Bivalirudin was used with or without loading dose, followed by
infusion at different ranges (lowest 0.1-0.2 mg/kg/h without loading dose; highest 0.5 mg/kg/h after loading dose). The strategies
for monitoring anticoagulation and optimal targets were dissimilar (activated partial thromboplastin time 45-60 seconds to 42-88
seconds; activated clotting time 180-200 seconds to 200-220 seconds; thromboelastography in 1 study). Conclusion: Bivalirudin
loading dose was not always used; infusion range and anticoagulation targets were different. In this systematic review, we discuss
the reasons for this variability. Larger studies are needed to establish the optimal approach with the use of bivalirudin for ECMO.

Keywords
cardiopulmonary bypass, ECMO, heparin-induced thrombocytopenia, HIT, direct thrombin inhibitor

Introduction
Extracorporeal membrane oxygenation (ECMO) can provide
cardiac and/or respiratory support even for extended periods.
It can be used for venovenous circulation (VV-ECMO) to pro-
vide adequate oxygenation and carbon dioxide removal in iso-
lated refractory respiratory failure of various causes or in a
venoarterial configuration (VA-ECMO) when support is
required for cardiac function.1-3 The ECMO circuits are shorter
than the traditional cardiopulmonary bypass (CPB) circuit and
are also heparin bonded, therefore requiring lower levels of
anticoagulation.4 Unfractionated heparin remains the anticoa-
gulant of choice because of easy titration and monitoring, low
costs, overall familiarity with its use, and the availability of an
antidote. Nonetheless, the use of unfractionated heparin or, less
commonly, of low-molecular-weight heparins5 and fondapar-
inux may trigger heparin-induced thrombocytopenia (HIT),6 a
potentially life-threatening immune disorder7,8 characterized
by declining platelet counts (5-14 days after heparin exposure)
in isolation (isolated HIT) or concurrently with thrombotic
complications (heparin-induced thrombocytopenia with
1
Department of Anesthesia and Intensive Care, IRCCS-ISMETT (Istituto
Mediterraneo per i Trapianti e Terapie ad alta specializzazione), Palermo, Italy
2
School of Anaesthesia and Intensive Care, University of Catania, Catania, Italy
3
Department of Anaesthesiology, Ruhr-University Bochum, University Hos-
pital Knappschaftskrankenhaus, Bochum, Germany
4
Critical Care Research Group, The Prince Charles Hospital and The University
of Queensland, Brisbane, Queensland, Australia
5
Department of Anaesthesiology and Intensive Care Medicine, Philipps University,
Marburg, Germany
6
Department of Anaesthesiology and Pain Therapy, Inselspital, Bern University
Hospital, Bern, Switzerland
7
Cardiothoracic Anaesthesia and Intensive Care, Manchester Royal Infirmary,
Central Manchester University Hospitals NHS Foundation Trust, Manchester
Academic Health Science Centre, University of Manchester, Manchester,
United Kingdom
Received March 15, 2016. Received revised May 11, 2016. Accepted
June 3, 2016.
Corresponding Author:
Filippo Sanfilippo, Department of Anesthesia and Intensive Care, IRCCS-
ISMETT (Istituto Mediterraneo per i Trapianti e Terapie ad alta specializza-
zione), Via Tricomi 5, Palermo 90127, Italy.
Emails: filipposanfi@yahoo.it; fgsanfilippo@ismett.edu
Sanfilippo et al
Figure 1. Simplified coagulation cascade and action of anticoagulants.
AT indicates antithrombin; LMWH, low-molecular-weight heparins;
TF, tissue factor.
thrombosis syndrome [HITT]).9,10 An overall incidence of
2.6% of immune-mediated HIT in patients treated with unfrac-
tionated heparin was found by a meta-analysis,11 but its inci-
dence in patients exposed to long-term anticoagulation (ECMO
and ventricular assist devices) is higher.12,13 Moreover, throm-
bocytopenia is very common in ECMO patients, and differen-
tiating between HIT and other causes (platelet activation or
dilution, excessive bleeding) is challenging.
Another possible issue with anticoagulation for extracorporeal
circuits is represented by the occurrence of heparin resistance
(HR), commonly caused by low antithrombin III levels. This
other challenge is often managed with supplementary heparin and
fresh frozen plasma or antithrombin III concentrate.14
In the case of HIT/HITT or of HR during ECMO, very lim-
ited evidence is available for alternative strategies for anticoa-
gulation. Bivalirudin, a direct thrombin inhibitor (DTI), does not
require antithrombin III (Figure 1) and has additional activity
against thrombin-mediated platelet activation; it is currently
approved by the Food and Drug Administration for use during
percutaneous coronary intervention/angioplasty.15 It has gradu-
ally become an accepted alternative in patients requiring antic-
oagulation in the setting of HIT/HITT, HR, and thrombosis.
Bivalirudin is also recommended as a first-line strategy for
intraoperative anticoagulation in patients with HIT/HITT requir-
ing cardiac surgery,16-18 and it is widely used during percuta-
neous coronary interventions,19-21 although a recent large trial
has not shown improved outcomes as compared with unfractio-
nated heparin.22 However, little is known about the use of biva-
lirudin during ECMO. This systematic review aims at evaluating
dosing effects and complications of bivalirudin-based anticoa-
gulation for patients treated with VV- or VA-ECMO.
Methods
This systematic review was based on the findings of a web-based
literature search conducted using the National Health Service
313
Table 1. ‘‘PICOS’’ Approach for Selecting Clinical Studies in the
Systematic Search.
PICOS
1 Participants Patients undergoing extracorporeal membrane
oxygenation (ECMO)
2 Intervention Both venovenous and venoarterial ECMO
anticoagulated with bivalirudin
3 Comparison Comparison with other anticoagulation strategy only
if available
4 Outcomes Optimal dose of bivalirudin, efficacy of
anticoagulation and anticoagulation-related
complications
5 Study design Prospective and retrospective clinical studies; case
series and case reports
Library Evidence tool. A further manual search was conducted
independently by 2 authors (F.S. and M.O.M.), exploring the list
of references of the findings of the systematic search. The
approach suggested by the Preferred Reporting Items for Sys-
tematic Reviews and Meta-Analyses (PRISMA) statement for
reporting systematic reviews and meta-analyses was followed.23
The primary purpose of this systematic review was to estab-
lish the current knowledge and experience with bivalirudin-
based anticoagulation and whether bivalirudin represents a safe
alternative method of anticoagulation during ECMO. A sec-
ondary end point of the review was to investigate an optimal
dosing of bivalirudin for ECMO.
Inclusion criteria were prespecified according to the Patients -
Intervention - Comparison - Outcome - Study (PICOS) approach
(Table 1). Since we expected no randomized controlled studies,
we a priori decided to include nonrandomized prospective and
retrospective clinical studies. The computerized search included
the 2 most significant health-care databases, MEDLINE
(PubMed) and EMBASE, from inception until January 30th,
2016. The core search was structured by the combination of
terms obtained from the following 2 groups. The first one
included the following in alphabetical order: ‘‘bivalirudin’’ and
‘‘direct thrombin inhibitor.’’ The second group consisted of the
following: ‘‘extracorporeal membrane oxygenation,’’
‘‘ECMO,’’ ‘‘extracorporeal life support,’’ and ‘‘ECLS.’’ The
search strategy was limited to clinical (human) studies, and it
is summarized in the Supplemental Digital Content, Appendix 1.
Two authors (F.S. and M.O.M.) independently searched
these databases. Duplicates were initially filtered through auto-
mated software function and afterward screened manually by 3
authors (F.S., D.M.M., and M.O.M.). Study selection for deter-
mining the eligibility for inclusion in the systematic review and
data extraction from the selected studies were performed inde-
pendently by 3 reviewers (F.S., D.M.M., and M.O.M.). Discor-
dances were resolved by involving the other reviewers (J.F.F.,
C.S., G.E., and S.A.) and/or by consensus.
No language restrictions were applied. Findings retrieved
from EMBASE as conference abstract are reported only if
published after February 2012 to allow a reasonable time for
multiple peer-reviewed process. Book chapters, reviews,
314
Figure 2. Flow diagram of the systematic search.
editorials, and letters to editor were excluded from the qualita-
tive synthesis. Grading of the studies and assessing the risk of
bias were performed by 3 authors (F.S, S.A., and M.O.M.).
Results
The literature search with the above-mentioned criteria pro-
duced 325 total findings, 101 on MEDLINE and 224 on
EMBASE; 24 duplicates were removed via automatic software,
leaving a total of 301 publications.
We initially excluded 289 findings as judged not relevant to
our search target or for their design (letter to the editor), leaving
potentially 12 findings. We identified manually 3 further dupli-
cates, and the manual search did not add further findings. The
remaining 9 findings (7 on PubMed and 2 on EMBASE) were
initially included in the qualitative synthesis. However, 1 case
report (conference abstract on EMBASE) was excluded during
the qualitative synthesis for lack of valuable information,24
leaving finally 8 included studies, 1 case control/cohort
study,25 1 retrospective chart review,26 1 case series,27 and 5
case reports7,12,24,28-30 (flow diagram of the search according to
PRISMA statement is shown in Figure 2). All articles were
published between 2007 and 2015.
The aggregation of available publications displays a limited
number of 58 patients; 24 (41%) of them belonging to the
pediatric population. A total of 18 patients received heparin
(control groups in the 2 larger studies), whereas 40 received
bivalirudin (treatment groups/cases). Among these 40 patients,
almost half of them were pediatric cases (n ¼ 19, 47%), and the
reasons for using bivalirudin are reported as follows: n ¼ 9
patients had HIT, n ¼ 6 had HR, n ¼ 4 had unstable activated
clotting time (ACT), and n ¼ 2 had clotting while on heparin.
The reason for bivalirudin anticoagulation is not reported for 6
of 10 patients in the study by Pieri et al.25 Finally, the remain-
ing 13 patients anticoagulated with bivalirudin were included
from the study by Ranucci et al26 following a change in the
internal policy of anticoagulation at their hospital. The
Journal of Intensive Care Medicine 32(5)
important outcomes, dosage used, and anticoagulation targets
adopted are shown in Table 2.
Results of the risk of bias assessment are shown in Figure 3.
Allocation bias, performance bias, and determination bias are
high throughout the studies as expected, due to the study design
as case reports. In addition, no sufficient information was pro-
vided in the case/control studies25,26 how data were handled
after grouping and whether the authors took any precaution to
minimize performance or determination bias caused by the
investigators.
Discussion
The systematic literature search revealed a limited number of 3
studies and 5 case reports on the use of bivalirudin in patients
on ECMO, for a total of 58 patients only, 24 of them belonging
to the pediatric population. The 2 studies comparing patients
anticoagulated with bivalirudin versus heparin strategy25,26
reported no difference in the complication rate. However, one
of these studies26 showed higher blood losses and transfusion
requirements in the heparin group. The authors hypothesized
the absence of heparin-mediated platelet consumption as a pos-
sible mechanism to explain reduced bleeding and transfusion in
the bivalirudin group. Nonetheless, this remains speculative,
and it is challenging to ascertain the safety of bivalirudin on
ECMO from just 2 retrospective studies with a small number of
patients.
Bivalirudin Dose
In our systematic review, we found that authors have reported
very different doses of bivalirudin during ECMO, together with
a high variability in the protocol of anticoagulation with biva-
lirudin. Indeed, bivalirudin was used both with and without
loading dose, followed by continuous infusion at variable
ranges. Excluding the pediatric case using very high doses of
bivalirudin justified by the concomitant plasma exchange,30
Sanfilippo et al 315

Table 2. Grading of Manuscripts, Dosing and Anticoagulation, Patient-Important Outcomes.a

First Author, Dosing and Anticoagulation

Year Study Design ECMO Patients Included (#) Targets Important Outcomes

1 Pieri,25 2013 Case–control 20 Adult patients on VV (10) or No bivalirudin loading; median
retrospective VA-ECMO (10); 10 in the H
study group (randomly selected after
stratification); and 10 in the B 0.05) mg/kg/h; without
group (5 per each ECMO type)
2 Ranucci,26 Case–control 21 Patients (11 adults and 10
2011 retrospective pediatrics) undergoing
study postcardiotomy VA-ECMO.
Eight patients in the H group (5 seconds, every 4 hours), aPTT
pediatrics) and 13 patients in
the B group (5 pediatrics)
3 Nagle,27 2013 Case series 12 Pediatrics (8 of them neonatal), Four patients received a loading
362 + 752-day-old; VV-ECMO
(n ¼ 3) or VA-ECMO (n ¼ 9).
n ¼ 5 HR, n ¼ 4 unstable ACTs,
n ¼ 2 clotting on heparin, and
n ¼ 1 HIT
4 Preston,30 Case report An 8-year-old patient on VV- Bivalirudin infusion: 1.2 to 1.8
2015 ECMO as bridge to lung
transplantation; HIT and
necessitating plasma exchange
for allosensitization
5 Jyoti,7 2014 Case report A 54-year-old patient with H1N1 No bivalirudin loading; average
infection on VV-ECMO;
antithrombin deficiency, HR,
and thrombosis
infusion dose: with
hemofiltration 0.041 (0.028-
hemofiltration 0.028 (0-0.041)
mg/kg/h. Target aPTT was 45
to 60 seconds
No bivalirudin loading; initial
infusion dose 0.03 to 0.05
mg/kg/h. Targets: ACT (160-180
(50-80 seconds, every 12
hours), and R time at TEG
(12-30 minutes, every 8 hours)
values
(median 0.1 mg/kg, range 0.04-
0.14). Initial dose: 0.05 to 0.3
mg/kg/h. Maintenance dose:
0.045 to 0.48 mg/kg/h aPTT for
monitoring, target not specified
mg/kg/h associated to bivalirudin
boluses (0.75-1.5 mg/kg) during
2 cycles of plasma exchange.
aPTT 60 to 80 seconds
infusion dose 0.1 to 0.2 mg/kg/
h. ACT target 200 to 220
seconds
The H group had significantly
more episodes of aPTT
variation >20%.
Nonsignificantly higher number
of episodes with aPTT >80
seconds and anticoagulant dose
adjustment in the H group.
Nonsignificant increase of
(minor þ major) bleeding in the
H group (6 vs 3 in B).
Nonsignificantly higher
mortality in the H group (50%
vs 40% in B)
The B group: significantly longer
ACTs, aPTTs, and R time (in
the target range). The H group:
higher blood losses (51 vs 16
mL/kg/d) and administration of
platelet (33 vs 3 mL/kg/d), FFP
(12 vs 6 mL/kg/d), and
antithrombin (13 vs 7 mL/kg/d).
No difference in platelet counts
or thromboembolic events.
Mortality on ECMO 62.5% in
the H group (n ¼ 5 of 8), 31% in
the B group (n ¼ 4 of 13). One
patient in the H and 6 in the B
group died after ECMO
weaning. Lower costs for the
pediatric ECMO in the B group
(€312 vs €760 per day).
Nonsignificant cost reduction
in the adults (€1807 vs €3313
per day)
Mortality was 33% while on
ECMO, 58% hospital mortality.
No intracranial hemorrhage
detected on ultrasound. Two
patients required chest
reexploration. Three patients
required recombinant factor
VIIa for operative procedures.
Estimated costs for B: US$13.7/
kg/d (based on median infusion
0.16 mg/kg/h) compared to
US$0.5 kg/d with the H therapy
(based on rate of 50 U/kg/h)
Less fluctuation of aPTT when
using FFP alone for volume
exchange rather than when
using a mixture containing 25%
albumin
175 mL/d of packed red blood cell
transfused. Platelet count
stable, no complications like
bleeding or thrombosis were
observed. ECMO weaned after
23 days and discharge home
4 weeks later
(continued)
316 Journal of Intensive Care Medicine 32(5)

Table 2. (continued)

First Author, Dosing and Anticoagulation

Year Study Design ECMO Patients Included (#) Targets Important Outcomes

6 Pollak,28 2011 Case report A 5-day-old neonatal patient with Bivalirudin loading 0.4 mg/kg;
left diaphragmatic hernia on
VA-ECMO; HIT with small-
vessel arterial thrombosis
7 Pappalardo,29 Case report A 71-year-old patient
2009 postcardiotomy RV failure on
VA-ECMO; HIT
8 Koster,12 Case report A 40-year-old patient
2007 postcardiotomy (type A aortic
dissection) biventricular failure
on VA-ECMO; HIT
initial infusion 0.15 mg/kg/h.
Infusion had to be increased up
to 1.6 mg/kg/h (multiorgan
failure). ACT target 180 to 200
seconds
Bivalirudin loading 0.5 mg/kg;
infusion of 0.05 to 0.15 mg/kg/h.
aPTT target 42 88 seconds
Bivalirudin loading for ECMO was After 8 days, patient received an
0.5 mg/kg; followed by
continuous infusion 0.5 mg/kg/
h. ACT target 200 to 220
seconds
Diaphragmatic hernia repair
performed on ECMO. Ten
platelets transfusion (4 of the H
group in 6 days and 6 of the B
group in 15 days). Patient died
after 21 days of ECMO
(multiorgan failure)
HIT persisted for the duration of
ECMO probably due to
heparin-coated tubing. Atrial
thrombus and oxygenator
thrombi formation. ECMO
weaned after 6 days.
Discharged home after 1 month
RVAD on the B therapy
(additional bolus and infusion
up to 1 mg/kg/h) was weaned
from VA-ECMO. On RVAD,
anticoagulation was continued
on argatroban (0.05 mg/kg/h).
Weaned successfully from
RVAD after 6 weeks

Abbreviations: ACT, activated clotting time; aPTT, activated partial thromboplastin time; CABG, coronary artery bypass graft; CDH, congenital diaphragmatic
hernia; ECMO, extracorporeal membrane oxygenation; FFP, fresh frozen plasma; HIT, heparin-induced thrombocytopenia; HR, heparin resistance; MVR, mitral
valve replacement; RVAD, Right Ventricular Assist Device; TEG, thromboelastography; VA, venoarterial; VV, venovenous.
a
The manuscripts are presented in descending grade of evidence with regard to bivalirudin (B) and heparin (H) anticoagulation for ECMO patients.

Figure 3. Assessment of the risk of bias for the studies included in the
systematic review. Case series and case report were not assessed for
risk of bias since they have no control group.
Nagle et al27 reported the highest initial infusion of bivalirudin
(0.48 mg/kg/h) in a pediatric population, with a loading dose
given to 4 patients only due to either thrombus formation or
subtherapeutic activated partial thromboplastin time (aPTT)
values. The 2 largest studies did not use a loading dose and
reported lower ranges of bivalirudin infusion, from 0.028 to
0.05 mg/kg/h.26,31 In 1 case report of a patient with HR, biva-
lirudin infusion was performed at higher doses (0.1-0.2 mg/
kg/h) without a loading dose,7 whereas in the other 3 case
reports, a loading dose was used (range: 0.4-0.5 mg/kg)
and followed by 10-fold variability in infusion range
(0.05-0.5 mg/kg/h).12,28,29 The observed interpatient variabil-
ity is reinforced comparing the case reported by Jyoti et al7
(0.1-0.2 mg/kg/h without loading dose) that used less than
half the dose than the case presented by Koster et al 12
(0.5 mg/kg/h after a loading dose), aiming at the same ACT
target range. The remarkable dose difference is also in line
with the findings of the largest pediatric series27 of bivalirudin
during ECMO that used an about 10-fold variable bivalirudin
dose (range: 0.045-0.48 mg/kg/h).
In this context of dose variability, a couple of comments
should be made. First, bivalirudin metabolism certainly plays
a key role, being the molecule cleared by a combination of
renal mechanisms (20% in patients with unimpaired renal func-
tion) and proteolytic cleavage.32 In a large single-center retro-
spective study of 135 patients treated with bivalirudin for HIT,
patients with renal dysfunction required lower bivalirudin
doses to achieve target aPTT. The authors found that in patients
with renal dysfunction, significantly lower doses of bivalirudin
were required for achieving the target aPTT (creatinine
Sanfilippo et al
clearance >60 mL/min: bivalirudin 0.13 mg/kg/h vs 30-60
mL/min: 0.08 mg/kg/h and vs <30 mL/min: 0.05 mg/kg/h;
P < .001)33 However, patients on continuous renal replacement
therapy required higher bivalirudin doses (0.07 mg/kg/h) than
patients with creatinine clearance <30 mL/min and not receiv-
ing dialysis. Similarly, although nonsignificant, Pieri et al25
found that bivalirudin clearance was affected by hemofiltra-
tion, and patients on renal replacement therapy required higher
doses as compared to patients who did not (0.041 vs 0.028
mg/kg/h). It seems therefore reasonable to adjust the bivaliru-
din doses in sicker patients with compromised renal clearance
according to the use of dialysis, rather than waiting for the
accumulation of the drug, which may increase bleeding risks,
or for subtherapeutic anticoagulation targets.
Second, in consideration of the short half-life of bivalirudin
(25 minutes),34 the use of a loading dose should be carefully
evaluated. Indeed, a steady state is achieved in most patients
within a couple of hours after bivalirudin infusion and a loading
dose is probably unnecessary unless in the presence of a largely
subtherapeutic range and/or in the context of suspected throm-
bosis. Moreover, in case of high risk of bleeding such as in
postcardiotomy VA-ECMO, it could be reasonable to avoid a
loading dose. On the basis of the information retrieved, we
believe a loading dose was appropriately given by Pollak
et al28 in a patient with arterial thrombosis and by Koster
et al12 in a patient with a mechanical aortic valve conduit (after
type A aortic dissection). Less strong seems the indication for a
loading dose in a case of HIT and postcardiotomy right ven-
tricular failure after biological mitral valve replacement and
coronary grafts reported by Pappalardo et al,29 since biological
valves have a lower risk of thrombosis. On the other side, one
would have expected a loading dose in the case discussed by
Jyoti et al7 because the authors reported evidence of thrombosis
in the oxygenator of the ECMO circuit.
Bivalirudin and Anticoagulation Targets
When looking at bivalirudin doses used, the choice of moni-
toring is confounding. As highlighted in Table 2, ACT targets
varied from 180 to 200 seconds to 200 to 220 seconds, aPTT
from 45 to 60 seconds to 42 to 88 seconds, and in 1 study, a
more complex monitoring including ACT (160-200 seconds),
aPTT (50-80 seconds), and thromboelastography (R time 12-30
minutes) was used as well.
All these goals for anticoagulation seem reasonable and
appropriate; importantly, it is very likely that each center has
developed their guidelines and preferences based on experi-
ence matured throughout many years. It is also worth noting
that in the study by Pieri and colleagues,25 the bivalirudin
group had lower incidence of significant aPTT fluctuations as
compared to the heparin group.
Interestingly, Ranucci et al have recently shown the pres-
ence of a ‘‘heparin-like effect’’ in 41 postcardiotomy VA-
ECMO patients anticoagulated solely with bivalirudin
(R-times with thromboelastography significantly shorter in
tests with vs without heparinase). Patients with ‘‘heparin-like
317
effect’’ (56.1% of the study population) were more likely to
have sepsis (30% vs 5.6%) possibly as a result of the release of
heparinoids from the glycocalyx and the mast cells due to
sepsis or of the systemic inflammatory response. Therefore,
more sick patients could be exposed to further variation in
anticoagulation targets as result of their critical illness.35 How-
ever, this hypothesis should be confirmed as sicker patients are
also more likely to have a deranged renal function and therefore
bivalirudin clearance.
Bivalirudin and Thrombosis
The short half-life of bivalirudin (25 minutes) has major impli-
cations for patients on extracorporeal circulation, as any region
of stasis of blood in the system/patient, or areas of blood being
disconnected from the systemic blood flow, may lead to a
critical decrease in the regional bivalirudin concentrations and
may cause thrombus formation.36,37 Ranucci invoked for a
word of caution, given the pharmacological profile of bivalir-
udin and its short half-life.31
The use of bivalirudin for CPB and ECMO is a feasible
option, but blood stagnation in the circuit must be avoided for
obvious reasons. The ECMO circuit is devoid of a reservoir,
and therefore, blood stagnation is usually not a circuit-related
problem. Nonetheless, in the absence of effective cardiac con-
traction, the heart chambers may act as a ‘‘natural reservoir,’’
which entails blood stagnation and risk for spontaneous intra-
cardiac thrombosis. To avoid this condition, while maintaining
a partial VA-ECMO support, facilitating/inducing some degree
of cardiac contraction may be reasonable. Intracardiac throm-
bus formation during ECMO with heparin anticoagulation has
been described as well, but the pharmacokinetic properties of
heparin (half-life: 1-2 hours) may limit the risk of thrombus
formation due to blood stagnation.38
Pappalardo and colleagues29 reported on a patient with acute
HIT who underwent ECMO with a heparin-coated system and
bivalirudin infusion. In this patient, the HIT reaction continued
with concomitant thromboembolic event despite systemic
anticoagulation, and this was interpreted as caused by the
release of heparin from the circuit. However, no conclusion
can be drawn on the effects of heparin-bonded tubing for
patients with HIT based on a single case. Unbonded tubing
and/or higher anticoagulation targets may be considered on
individual basis.
Alternatives to Bivalirudin
Among the DTIs, bivalirudin has the shortest half-life and
potentially the best pharmacological profile for use during
ECMO, especially given the broad experience with bivalirudin
in patients undergoing cardiac surgery and CPB.36,39 The short
half-life of bivalirudin may also be advantageous compared to
other DTIs such as argatroban (39-51 minutes) and lepirudin
(78 minutes) for titration of anticoagulation.40,41 Moreover,
argatroban and lepirudin may accumulate in hepatic and renal
dysfunction, respectively, potentially leading to derangements
318
of anticoagulation target and thus higher incidence of
complications.
Although data supporting the reversal of DTIs are limited,
reversal of bivalirudin using recombinant factor VII (rFVIIa)
has been described,42 and the use of rFVIIa in intractable
hemorrhage in patients receiving ECMO seems promising.43,44
Among other alternatives to heparin anticoagulation, danapar-
oid and fondaparinux have a half-life of approximately
24 hours and therefore may be suboptimal for ECMO support.
Moreover, the available experience with these 2 drugs is lim-
ited to 1 case report each,45,46 and the use of fondaparinux in
case of HIT/HITT is questionable in the presence of
alternatives.
Limitations
The main limitations of this systematic review include the
retrospective nature of all of the evidence retrieved, the small
number of studies and patients included, variable patient’s
characteristics, bivalirudin dosage, and target parameters for
anticoagulation. Importantly, the retrospective design made
unreliable capturing the occurrence of complication such as
hemorrhage and thrombosis. The blood product requirements
and cannula site bleeding cannot be determined from the pres-
ent literature. Thrombus formation within the circuit cannot be
excluded, and circuit life span was not reported.
Conclusion
Bivalirudin seems to be the best viable option for patients on
ECMO who present contraindications to or ineffectiveness of
heparin anticoagulation. Bivalirudin has been used both with
and without a loading dose, and as for heparin anticoagulation,
it requires strict and continuous dose adjustment according to
prespecified targets. Prospective larger randomized controlled
studies are necessary to fully determine both the safety and
efficacy for the use of bivalirudin during ECMO.
Declaration of Conflicting Interests
The author(s) declared no potential conflicts of interest with respect to
the research, authorship, and/or publication of this article.
Funding
The author(s) received no financial support for the research, author-
ship, and/or publication of this article.
Supplemental Material
The online appendix is available at http://journals.sagepub.com/doi/
suppl/10.1177/0885066616656333.
References
1. Combes A, Brodie D, Bartlett R, et al; International ECMO Net-
work (ECMONet). Position paper for the organization of extra-
corporeal membrane oxygenation programs for acute respiratory
failure in adult patients. Am J Respir Crit Care Med. 2014;190(5):
488-496.
Journal of Intensive Care Medicine 32(5)
2. Asmussen S, Maybauer DM, Fraser JF, et al. Extracorporeal
membrane oxygenation in burn and smoke inhalation injury.
Burns. 2013;39(3):429-435.
3. Peek GJ, Clemens F, Elbourne D, et al. CESAR: conventional
ventilatory support vs extracorporeal membrane oxygenation
for severe adult respiratory failure. BMC Health Serv Res.
2006;6:163.
4. Ailawadi G, Zacour RK. Cardiopulmonary bypass/extracorporeal
membrane oxygenation/left heart bypass: indications, techniques, and
complications. Surg Clin North Am. 2009;89(4):781-796, vii-viii.
5. Cohen M. Heparin-induced thrombocytopenia and the clinical use
of low molecular weight heparins in acute coronary syndromes.
Semin Hematol. 1999;36(1 suppl 1):33-36.
6. Warkentin TE. Fondaparinux: does it cause HIT? Can it treat
HIT? Expert Rev Hematol. 2010;3(5):567-581.
7. Jyoti A, Maheshwari A, Daniel E, Motihar A, Bhathiwal RS,
Sharma D. Bivalirudin in venovenous extracorporeal membrane
oxygenation. J Extra Corpor Technol. 2014;46(1):94-97.
8. Greinacher A. Clinical Practice. Heparin-induced thrombocyto-
penia. N Engl J Med. 2015;373(3):252-261.
9. Warkentin TE, Sheppard JA. Testing for heparin-induced throm-
bocytopenia antibodies. Transfus Med Rev. 2006;20(4):259-272.
10. Lee GM, Arepally GM. Heparin-induced thrombocytopenia.
Hematology Am Soc Hematol Educ Program. 2013;2013:
668-674.
11. Martel N, Lee J, Wells PS. Risk for heparin-induced thrombocy-
topenia with unfractionated and low-molecular-weight heparin
thromboprophylaxis: a meta-analysis. Blood. 2005;106(8):
2710-2715.
12. Koster A, Weng Y, Böttcher W, Gromann T, Kuppe H, Hetzer R.
Successful use of bivalirudin as anticoagulant for ECMO in a
patient with acute HIT. Ann Thorac Surg. 2007;83(5):1865-1867.
13. Warkentin TE, Greinacher A, Koster A. Heparin-induced throm-
bocytopenia in patients with ventricular assist devices: are new
prevention strategies required? Ann Thorac Surg. 2009;87(5):
1633-1640.
14. Beattie GW, Jeffrey RR. Is there evidence that fresh frozen
plasma is superior to antithrombin administration to treat heparin
resistance in cardiac surgery? Interact Cardiovasc Thorac Surg.
2014;18(1):117-120.
15. Angiomax [Package insert]. Parsippany N, Medicines Company;
2000. Web site. https://web.archive.org/web/20120426012238/
http://angiomax.com/Downloads/Angiomax_PI_2010_PN1601-
12.pdf. Accessed June 16, 2016.
16. Aouifi A, Blanc P, Piriou V, et al. Cardiac surgery with cardio-
pulmonary bypass in patients with type II heparin-induced throm-
bocytopenia. Ann Thorac Surg. 2001;71(2):678-683.
17. Koster A, Dyke CM, Aldea G, et al. Bivalirudin during car-
diopulmonary bypass in patients with previous or acute
heparin-induced thrombocytopenia and heparin antibodies:
results of the CHOOSE-ON trial. Ann Thorac Surg. 2007;
83(2):572-577.
18. Dyke CM, Smedira NG, Koster A, et al. A comparison of bivalir-
udin to heparin with protamine reversal in patients undergoing
cardiac surgery with cardiopulmonary bypass: the EVOLUTION-
ON study. J Thorac Cardiovasc Surg. 2006;131(3):533-539.
Sanfilippo et al
19. Stone GW, McLaurin BT, Cox DA, et al; ACUITY Investigators.
Bivalirudin for patients with acute coronary syndromes. N Engl J
Med. 2006;355(21):2203-2216.
20. Stone GW, Witzenbichler B, Guagliumi G, et al; HORIZONS-
AMI Trial Investigators. Bivalirudin during primary PCI in
acute myocardial infarction. N Engl J Med. 2008;358(21):
2218-2230.
21. Kushner FG, Hand M, Smith SC Jr, et al. 2009 Focused updates:
ACC/AHA guidelines for the management of patients with ST-
elevation myocardial infarction (updating the 2004 guideline
and 2007 focused update) and ACC/AHA/SCAI guidelines on
percutaneous coronary intervention (updating the 2005 guide-
line and 2007 focused update) a report of the American College
of Cardiology Foundation/American Heart Association Task
Force on Practice Guidelines. J Am Coll Cardiol. 2009;54(23):
2205-2241.
22. Valgimigli M, Frigoli E, Leonardi S, et al; MATRIX Investiga-
tors. Bivalirudin or unfractionated heparin in acute coronary syn-
dromes. N Engl J Med. 2015;373(11):997-1009.
23. Liberati A, Altman DG, Tetzlaff J, et al. The PRISMA statement
for reporting systematic reviews and meta-analyses of studies that
evaluate health care interventions: explanation and elaboration.
J Clin Epidemiol. 2009;62(10):e1-e34.
24. Sell RL, Patel K, Crane T. Bivalirudin use in heparin resis-
tance patient for extracorporeal membrane oxygenation: a
review and case study. J Extra Corpor Technol. 2009;41/
1(a17).
25. Pieri M, Agracheva N, Bonaveglio E, et al. Bivalirudin versus
heparin as an anticoagulant during extracorporeal membrane oxy-
genation: a case-control study. J Cardiothorac Vasc Anesth. 2013;
27(1):30-34.
26. Ranucci M, Ballotta A, Kandil H, et al; Surgical and Clin-
ical Outcome Research Group. Bivalirudin-based versus
conventional heparin anticoagulation for postcardiotomy
extracorporeal membrane oxygenation. Crit Care. 2011;
15(6):r275.
27. Nagle EL, Dager WE, Duby JJ, et al. Bivalirudin in pediatric
patients maintained on extracorporeal life support. Pediatr Crit
Care Med. 2013;14(4):e182-e188.
28. Pollak U, Yacobobich J, Tamary H, Dagan O, Manor-Shulman O.
Heparin-induced thrombocytopenia and extracorporeal mem-
brane oxygenation: a case report and review of the literature.
J Extra Corpor Technol. 2011;43(1):5-12.
29. Pappalardo F, Maj G, Scandroglio A, Sampietro F, Zangrillo A,
Koster A. Bioline heparin-coated ECMO with bivalirudin antic-
oagulation in a patient with acute heparin-induced thrombocyto-
penia: the immune reaction appeared to continue unabated.
Perfusion. 2009;24(2):135-137.
30. Preston TJ, Dalton HJ, Nicol KK, Ferrall BR, Miller JC, Hayes D
Jr. Plasma exchange on venovenous extracorporeal membrane
oxygenation with bivalirudin anticoagulation. World J Pediatr
Congenit Heart Surg. 2015;6(1):119-122.
31. Ranucci M. Bivalirudin and post-cardiotomy ECMO: a word of
caution. Crit Care. 2012;16(3):427.
319
32. Robson R, White H, Aylward P, Frampton C. Bivalirudin phar-
macokinetics and pharmacodynamics: effect of renal function,
dose, and gender. Clin Pharmacol Ther. 2002;71(6):433-439.
33. Tsu LV, Dager WE. Bivalirudin dosing adjustments for reduced
renal function with or without hemodialysis in the management of
heparin-induced thrombocytopenia. Ann Pharmacother. 2011;
45(10):1185-1192.
34. Warkentin TE, Greinacher A, Koster A. Bivalirudin. Thromb
Haemost. 2008;99(5):830-839.
35. Ranucci M, Baryshnikova E, Isgro G, et al. Heparin-like effect in
postcardiotomy extracorporeal membrane oxygenation patients.
Crit Care. 2014;18(5):504.
36. Linkins LA, Dans AL, Moores LK, et al; American College of
Chest Physicians. Treatment and prevention of heparin-induced
thrombocytopenia: Antithrombotic Therapy and Prevention of
Thrombosis, 9th ed: American College of Chest Physicians
Evidence-Based Clinical Practice Guidelines. Chest. 2012;141(2
suppl):e495s-e530s.
37. Morshuis M, Boergermann J, Gummert J, Koster A. A modified
technique for implantation of the HeartWare left ventricular assist
device when using bivalirudin anticoagulation in patients with
acute heparin-induced thrombocytopenia. Interact Cardiovasc
Thorac Surg. 2013;17(2):225-226.
38. Grimaldi A, Ajello S, Scandroglio M, et al. Intracardiac clots
masked by extracorporeal membrane oxygenation venous can-
nula. J Cardiothorac Vasc Anesth. 2012;26(2):e13-e14.
39. Koster A, Buz S, Krabatsch T, Yeter R, Hetzer R. Bivalirudin
anticoagulation during cardiac surgery: a single-center experience
in 141 patients. Perfusion. 2009;24(1):7-11.
40. Argatroban [package insert]. Research Triangle Park N, GlaxoS-
mithKline; 2009. Web site. https://www.gsksource.com/pharma/
content/dam/GlaxoSmithKline/US/en/Prescribing_Information/
Argatroban/pdf/ARGATROBAN.PDF. Accessed June 16, 2016.
41. Refludan [package insert]. Montville N, Berlex; 2004. Web site.
https://www.accessdata.fda.gov/drugsatfda_docs/label/2006/
020807s011lbl.pdf. Accessed June 16, 2016.
42. Nagle EL, Tsu LV, Dager WE. Bivalirudin for anticoagulation
during hypothermic cardiopulmonary bypass and recombinant
factor VIIa for iatrogenic coagulopathy. Ann Pharmacother.
2011;45(9):e47.
43. Vavra KA, Lutz MF, Smythe MA. Recombinant factor VIIa to
manage major bleeding from newer parenteral anticoagulants.
Ann Pharmacother. 2010;44(4):718-726.
44. Repesse X, Au SM, Brechot N, et al. Recombinant factor VIIa for
uncontrollable bleeding in patients with extracorporeal membrane
oxygenation: report on 15 cases and literature review. Crit Care.
2013;17(2):r55.
45. Bauer C, Vichova Z, Ffrench P, et al. Extracorporeal membrane
oxygenation with danaparoid sodium after massive pulmonary
embolism. Anesth Analg. 2008;106(4):1101-1103.
46. Parlar AI, Sayar U, Cevirme D, Yuruk MA, Mataraci I. Successful
use of fondaparinux in a patient with heparin-induced thrombo-
cytopenia while on extracorporeal membrane oxygenation after
mitral valve redo surgery. Int J Artif Organs. 2014;37(4):344-347.