Available via license: CC BY-NC-ND 4.0
Content may be subject to copyright.
Apixaban removal during emergency surgery for
type A acute aortic dissection: a prospective cohort
study
Henriette Røed-Undlien, MDa, Nina H. Schultz, MD, PhDb,c, Inger M. Husebråten, MScd, Birgit M. Wollmann, PhDe,
Rupali R. Akerkar, PhDf, Espen Molden, PhDe,g, Erik K. Amundsen, MD, PhDh,i, Johannes L. Bjørnstad, MD, PhDa,d,*
Background: Acute type A aortic dissection (ATAAD) has a high mortality, and acute aortic repair is the only curative treatment. In
patients treated with factor Xa (FXa) inhibitors, the risk of severe disease-related complications such as cardiac tamponade and
hemodynamic shock must be balanced against the potential for severe perioperative bleeding. The aim was to study intraoperative
changes in plasma levels of the FXa inhibitor apixaban when using hemoadsorption during acute thoracic aortic repair.
Materials and methods: This is a single-center prospective cohort study. Eight apixaban-treated patients presenting with ATAAD
underwent acute thoracic aortic repair with intraoperative hemoadsorption with CytoSorb. Apixaban concentrations were measured
at the start of cardiopulmonary bypass (CPB) and after 5, 15, 30, 60, and 90 min of CPB, at CPB weaning, 30 min after CPB weaning
and 24 h postoperatively, using ultraperformance liquid chromatography–mass spectrometry (UPLC-MS).
Results: After 30 min of CPB with hemoadsorption, mean apixaban concentration ( ±SD) was reduced by 59% from 108 ( ±69) µg/l
to 44 ( ±20) µg/l (P=0.009). There was a further reduction to 37 ( ±17) µg/l at CPB weaning (P=0.008). Apixaban concentration
displayed an increase to 56 ( ±29) µg/l 24 h postoperatively (P=0.01). In-hospital mortality was 25%. The mean 24H chest tube
drainage volume was 621 ( ±136) ml.
Conclusion: Intraoperative hemoadsorption lowers apixaban levels in patients undergoing emergency surgery for ATAAD. Further
research is needed to determine its impact on perioperative bleeding complications and mortality.
Keywords aortic dissection, aortic surgery, apixaban, hemoadsorption, hemodilution
Introduction
In recent years, direct oral anticoagulants (DOACs) have gained
popularity for mitigating thromboembolic risk in patients with
atrial fibrillation, as well as for treating deep venous thrombosis
(DVT) and pulmonary embolism (PE). In comparison to war-
farin, treatment with DOACs requires no routine follow-up, and
they have proven superior to warfarin in several studies, both
regarding effectiveness and safety
[1–3]
. Among the DOACs, the
Factor Xa (FXa) inhibitor apixaban is the one most frequently
prescribed in Norway, with 48 reported users per 1000 people
between 60 and 80 years in 2021
[4]
. When in need of emergency
cardiovascular surgery, patients treated with FXa inhibitors seem
to be at greater risk of perioperative complications, prolonged
hospital stay, and increased mortality
[5,6]
.
Reversal of the anticoagulant effect of FXa inhibitors in case of
emergency surgery has been a debated issue over recent years. The
antidote andexanet alfa, which acts as a FXa decoy, is not
available in all countries. Additionally, it has been found to be
incompatible with the use of heparin during cardiopulmonary
bypass (CPB)
[7,8]
. Existing guidelines suggest pausing anti-
coagulant treatment for at least 48 h prior to high-risk surgery,
such as cardiovascular surgery
[9]
. In an acute case of aortic dis-
section type A (ATAAD), postponing surgery to await drug
a
Institute of Clinical Medicine, University of Oslo,
b
Research Institute for Internal Medicine, Oslo University Hospital,
c
Department of Hematology, Oslo University Hospital,
d
Department of Cardiothoracic Surgery, Oslo University Hospital,
e
Center for Psychopharmacology, Diakonhjemmet Hospital,
f
Department of Health Registries, Norwegian
Institute of Public Health, Bergen,
g
Department of Pharmacy, Section for Pharmacology and Pharmaceutical Biosciences, University of Oslo,
h
Department of Medical
Biochemistry, Oslo University Hospital and
i
Department of Life Sciences and Health, Oslo Metropolitan University, Oslo, Norway
Sponsorships or competing interests that may be relevant to content are disclosed at the end of this article.
*Corresponding author. Address: Department of Cardiothoracic Surgery, Oslo University Hospital, Postboks 4950 Nydalen, 0424 Oslo, Norway. Tel.: +47 900 409 96, +47 915 027 70.
E-mail: uxbjjb@ous-hf.no (J.L. Bjørnstad).
Supplemental Digital Content is available for this article. Direct URL citations are provided in the HTML and PDF versions of this article on the journal’s website,
www.lww.com/international-journal-of-surgery.
Published online 18 November 2024
Received 21 August 2024; Accepted 5 November 2024
Copyright © 2024 The Author(s). Published by Wolters Kluwer Health, Inc. This is an open access article distributed under the terms of the Creative Commons Attribution-Non
Commercial-No Derivatives License 4.0 (CCBY-NC-ND), where it is permissible to download and share the work provided it is properly cited. The work cannot be changed in
any way or used commercially without permission from the journal.
International Journal of Surgery (2024) 110:7782–7790
http://dx.doi.org/10.1097/JS9.0000000000002137
’
Prospective Cohort Study
7782
elimination may be complicated by hemodynamic instability,
organ dysfunction, and a substantial early mortality rate
[10,11]
.
The coagulation system is activated when blood is exposed to the
media layer of the aortic wall in ATAAD, often resulting in a
reduction in clotting factors and fibrinogen, further increasing the
risk of uncontrolled bleeding during surgery
[12]
. Thus, an FXa
inhibitor antidote suitable for use with cardiopulmonary bypass
is lacking. Combined with the high mortality rate associated with
ATAAD that rises for each hour surgery is delayed, and the risk of
complicating acute renal dysfunction that further reduces the
elimination of FXa inhibitors in these patients, it is important to
investigate more effective strategies for reversing the anti-
coagulant effects of FXa inhibitors in this population
[13]
.
In recent years, hemoadsorptive devices such as CytoSorb have
been studied for their abilities to remove several different ther-
apeutic drugs from the circulation. CytoSorb has received EU reg-
ulatory approval for the removal of the FXa inhibitor rivaroxaban
during cardiac surgery. However, the clinical evidence for this is
limited to two retrospective studies and a few case reports
[14–16]
.
In 2022, we published the results of an experimental study
demonstrating that CytoSorb removes apixaban from blood in an
in-vitro circuit
[17]
. However, clinical studies reporting changes in
apixaban concentration using CytoCorb during cardiac surgery
are limited to two case reports with rather low apixaban levels
preoperatively
[18,19]
. Therefore, it remains uncertain whether
intraoperative hemoadsorption is an effective method for rever-
sing the anticoagulant effects of FXa inhibitors in patients
requiring emergency cardiac surgery.
In our present observational study, we aimed to study changes
in apixaban concentration in patients undergoing emergency
surgery for ATAAD using CytoSorb hemoadsorption intra-
operatively. The secondary aims of this study were to describe
postoperative bleeding complications, 30-day mortality, and
thromboembolic complications.
Materials and methods
Patients admitted with ATAAD at Oslo University Hospital from
August 2021 to August 2023 were considered for inclusion. The
inclusion criteria were: minimum of 18 years of age, indication
for urgent surgery for proximal aortic disease, treatment with
apixaban for more than 4 weeks, and a measured concentration
>30 µg/l at the start of surgery, according to previously published
proposals for preoperative DOAC concentrations
[20,21]
. The
exclusion criterion was apixaban concentration <30 µg/l.
The study received approval from The Norwegian National
Research Ethics Committee (REK #204588). The study was
recommended by the data protection officer at Oslo University
Hospital (#20/27731). Three participants were included based on
presumed consent following postoperative death or cognitive
impairment. All other study participants signed an informed
consent. The work has been reported in line with the STROCSS
criteria
[22]
(Supplemental Digital Content 1, http://links.lww.
com/JS9/D551).
Study design
This is a single-center prospective cohort study. A CytoSorb
cartridge (CytoSorbents, Berlin, Germany) was integrated into
the extracorporeal circuit from the start of cardiopulmonary
bypass intraoperatively (Fig. 1). Tubes, reservoir, and oxygenator
were coated with a Cortiva bioactive surface (Medtronic,
Minneapolis, Minnesota, USA). The CPB circuit with roller
pumps (Stöckert S5 HLM Sorin, Munich, Germany) was primed
according to institutional standards with 1200 ml Ringer Acetate,
200 ml mannitol, and 10 000 IE Heparin. The perfusion started at
a temperature of 32°C and was rapidly lowered to a minimum of
20°C measured through the arterial line. Before connecting the
hemoadsorptive device to the CPB circuit, the cartridge was
rinsed and primed with 2000 ml of saline solution according to
the manufacturer’s recommendations
[23]
. Figure 1 demonstrates
the placement of the CytoSorb cartridge in the CPB circuit. The
blood was pumped actively through the adsorber at a rate of
300 ml/min, and circulation through the adsorber was also
maintained during circulatory arrest.
Blood collection and analyses
Intraoperative blood samples were collected in citrated tubes
from the arterial shunt at the start of CPB, then after 5, 15, 30, 60,
and 90 min of CPB, and then at weaning from CPB (time point
named “x”). Blood samples were then drawn from the arterial
line at 30 min after weaning (time point named “e”) and 24 h
after the start of surgery. The blood samples were double cen-
trifuged at 2500gfor 15 min x2. Platelet-poor plasma (PPP) was
collected and stored at −80°C. Apixaban concentration was
measured by ultraperformance liquid chromatography
high-resolution mass spectrometry (QExactive high-resolution
mass spectrometer, Thermo Fisher Scientific, Waltham,
Massachusetts, USA) at the Center for Psychopharmacology,
Diakonhjemmet Hospital, Oslo, Norway. Analyses of albumin
and IgG (Cobas c702, Roche Diagnostics, Basel, Switzerland)
were performed in citrated plasma from each time point at the
Department for Medical Biochemistry, Oslo University Hospital,
Oslo, Norway. Measurements of hematocrit (Sysmex XN,
Sysmex Europe, Germany) were recorded from patient charts
postoperatively.
Statistics
Statistical analyses were performed using STATA 18.0 for
Windows and StataCorp 2024. Stata Statistical Software: Release
18. College Station, TX: StataCorp LLC and GraphPad Prism
version 10.2.0 for Windows, GraphPad Software, San Diego,
California, USA, www.graphpad.com. Results are presented as
mean and standard deviations (SD). Prior to analysis, normality
assumptions were assessed through the Shapiro–Wilk test and
QQ plots. A repeated measures one-way ANOVA was employed
to scrutinize variations across distinct time points. Furthermore,
intergroup mean comparisons were conducted using a two-sided
paired Student’sttest.
HIGHLIGHTS
•Intraoperative hemoadsorption in eight apixaban-treated
patients with ATAAD.
•Apixaban concentration measurements using UPLC-MS
revealed a 59% concentration decline during the first
30 min of CPB with hemoadsorption.
•No reoperations due to heavy bleeding were reported.
Røed-Undlien et al. International Journal of Surgery (2024)
7783
Results
Eight patients were included in the study, comprising one female
and seven male individuals. A summary of preoperative patient
characteristics is presented in Supplementary Table S1
(Supplemental Digital Content 2, http://links.lww.com/JS9/
D552). Participants’ages ranged from 73 to 83 years, with a
mean age of 76.8 years. The mean preoperative EURO II score
was calculated to be 18.09. One patient exhibited neurological
symptoms before the operation. Seven patients presented with a
DeBakey type 1 dissection, and one patient with an intramural
hematoma and a type 2 dissection (Supplementary Table S1,
Supplemental Digital Content 2, http://links.lww.com/JS9/
D552)
[24,25]
.
An overview of the most important preoperative, intraopera-
tive, and postoperative data is displayed in Table 1. Two out of
eight patients (25%) died in hospital. There were no reoperations
due to bleeding, but a second pump run due to troublesome
hemostasis and a small tear in the aortic wall (Supplementary
Table S2, Supplemental Digital Content 3, http://links.lww.com/
JS9/D553). The mean postoperative chest tube drainage (CTD)
was 621 ( ±136) ml. Among the eight patients included in the
study, three experienced postoperative strokes (37.5%). Of these,
one had symptoms of stroke preoperatively. Six patients (75%)
received activated prothrombin complex (aPCC) preoperatively
or intraoperatively. Of these, two (33%) received two dosages.
Four patients (50%) received four-factor prothrombin complex
concentrate (PCC) and one (12.5%) received recombinant acti-
vated factor VII (rFVIIa). All eight patients were treated with
tranexamic acid (TXA) and fibrinogen was administered in seven
cases (87.5%) (Table 1).
The mean apixaban concentration at the start of extracorporeal
circulation (t0) was 108 ( ±69) µg/l. After 30 min of CPB with
hemoadsorption, mean apixaban concentration was reduced by
59%to44(±20) µg/l (P=0.009) (Fig. 2). Apixaban concentration
was further reduced to a mean of 37 ( ±17) µg/l at CPB weaning
(P=0.008), and then displayed a slight increase to 45 (±19) µg/l
30 min after CPB weaning (P=0.01), which corresponds to 41.7%
of the initial mean apixaban concentration at t0. Apixaban con-
centration further displayed an increase to 56 ( ±29) µg/l 24 h
postoperatively (P=0.01). Individual apixaban concentrations are
displayed in Supplementary Figure S1 (Supplemental Digital
Content 4, http://links.lww.com/JS9/D554). A summary of intrao-
perative and postoperative data is provided in Supplementary Table
S2 (Supplemental Digital Content 3, http://links.lww.com/JS9/
D553) and Supplementary Table S3 (Supplemental Digital Content
5, http://links.lww.com/JS9/D555).
Hemodilution is common during CPB. Hematocrit was mea-
sured in routine samples at different time points than the study-
specific samples, and the values were recorded from the patients’
charts. There was a 23% reduction in hematocrit values for the first
15 min of surgery (Supplementary Fig. S2, Supplemental Digital
Content 4, http://links.lww.com/JS9/D554). In the study samples,
Figure 1. The CytoSorb was preconnected to a side port of the recirculation line from the arterial side and returned to the venous reservoir. Figure partly generated
using Servier Medical Art, provided by Servier, licensed under a Creative Commons Attribution 3.0 unported license.
Røed-Undlien et al. International Journal of Surgery (2024)
7784
there was a 30% reduction in albumin and IgG during the first
5 min of surgery. To illustrate the effect of hemodilution, Figure 3
displays the relative changes in apixaban, albumin, and IgG con-
centration at each time point, calculated in percentage from t0.
There was a 73% increase in mean plasma creatinine from 97
(±22) µmol/l preoperatively to 168 (±60) µmol/l measured 36 h
postoperatively (P=0.01) (Table 1). All patients were switched
from apixaban to warfarin for the first 3 months postoperatively,
as they all needed continuous anticoagulant therapy due to atrial
fibrillation (Supplementary Table S1, Supplemental Digital
Content 2, http://links.lww.com/JS9/D552). This is a local rou-
tine after cardiac surgery to reduce bleeding risk in case of the
need for additional procedures, such as pleural/pericardial drai-
nage or rethoracotomy. This is done as a safety precaution, as the
anticoagulant effect of warfarin better can be monitored and
easily and rapidly be reversed with 4F-PCC
[26]
.
Discussion
This is the first observational study examining the changes in
apixaban concentration when using hemoadsorption during
emergency surgery for ATAAD. A decrease in the plasma con-
centration of apixaban was demonstrated. At termination of
CPB, the mean apixaban concentration was at 38% of the initial
mean concentration, increasing to 55% of baseline concentration
after 24 h postoperatively.
Mean apixaban concentration was reduced by 44%
alreadyduringthefirst 5 min of CPB with hemoadsorption
Table 1
An overview of relevant preoperative, intraoperative, and postoperative data (n=8).
Patient # 1 2 3 4 5 6 7 8
Sum [%/Mean (min,
max)]
Preoperative
Sex F M M M M M M M 12.5% F
Age 73 77 76 83 76 75 74 80 76.8 (73, 83)
Indication for OAC AF
(parox.)
AF (perm.) AF
(perm.)
AF
(pers.)
AF
(parox.)
AF
(pers.)
AF
(parox.)
AF
(parox.)
100% AF
Previous thoracic surgery Yes (Ravitch
x2)
No No No No Yes
(CABG)
No No 25%
Neurological symptoms No No No No No No Yes No 12.5%
Delayed surgery No No No No No Yes Yes Yes 37.5%
Type of aortic dissection (DeBakey) Type 1 Type 1 Type 1 Type 1 Type 1 Type 1 Type 1 Intramural hematoma,
type 2
87.5% type 1
Cardiac tamponade Yes No No Yes Yes No Yes Yes 62.5%
Creatinine (µmol/l) 99 68 99 100 141 80 110 80 97 (68, 141)
Apixaban concentration (µg/l) 249 150 130 106 78 60 64 30 108 (30, 249)
Intraoperative
Type of surgery SCG SCG
CABG
SCG SCG SCG SCG SCG SCG 87.5% single, 12.5%
double
Duration of surgery (min) 260 190 187 215 185 263 212 323 229 (185, 323)
Duration of CytoSorb (min) 104 117 188 143 120 167 138 176 144 (104, 176)
Fibrinogen Yes Yes Yes Yes Yes No Yes Yes 87.5%
TXA Yes Yes Yes Yes Yes Yes Yes Yes 100%
Coagulation factor concentrates PCC
rFVIIa
aPCC
(x2)
aPCC PCC
aPCC
aPCC aPCC
(x2)
aPCC
PCC
PCC 75% aPCC
50% PCC
12.5% rFVIIa
Transfusions (units)
Erythrocytes 12 0 0 4 3 7 5 3 4.3 (0, 12)
Plasma 8 4 2 6 4 4 8 9 5.6 (2, 9)
Thrombocytes 2 2 1 2 2 2 3 1 1.9 (1, 3)
Postoperative
Transfusions (units)
Erythrocytes 4 9 1 3 3 3 4 3 3.8 (1, 9)
Plasma 8 17 4 6 4 8 2 3 6.5 (2, 17)
Thrombocytes 2 1 1 1 0 0 0 0 0.6 (0, 2)
Creatinine (µmol/l) 193 253 105 131 220 115 222 108 168 (105, 253)
Stroke No Yes No No Yes No Yes No 37.5%
CTD 24 h (ml) 620 770 670 770 630 470 660 380 621 (380, 770)
Rethoracotomy No No No No No No No No 0%
Apixaban concentration at CPB weaning
(µg/l)
60 49 57 27 39 27 23 14 37 (14, 60)
In-hospital death Yes
(2 weeks)
Yes
(8 days)
No No No No No No 25%
Røed-Undlien et al. International Journal of Surgery (2024)
7785
(Fig. 2). As shown in Figure 3, there was also a 30% decline
in albumin and IgG during this period of time. In our pre-
viously published experimental study, we found no significant
decrease in apixaban concentrations during the first 5 min of
hemoadsorption with CytoSorb
[17]
. Hence, we attribute most
of the early observed decrease in apixaban concentration to
be related to hemodilution, which is well known to occur
with the use of CPB
[27,28]
. This is supported by the transient
decline in albumin and IgG levels as well as hematocrit
(Supplementary Fig. S2, Supplemental Digital Content 4,
http://links.lww.com/JS9/D554). As illustrated in Figure 3,
there were minimal fluctuations in albumin levels from t5
onward throughout the surgery, while the apixaban con-
centration continued to diminish. Thus, we suggest that the
declineinapixabanconcentrationfromt5onwardcouldbe
attributable to the adsorptive device rather than hemodilu-
tion. Similarly, we believe that the increased apixaban con-
centration seen after 24 h was partly related to postoperative
hemoconcentration and drug redistribution
[29]
.Toour
knowledge, this is the first study to consider the effects of
hemodilution in addition to the adsorptive effects of
CytoSorb on drug concentrations.
As discussed by Connolly et al.
[30]
, following the ANNEXA-4
trial, reducing anti-factor Xa activity has not been found to
correlate well with hemostatic efficacy. Routine monitoring of
plasma concentrations of FXa inhibitors is not recommended, as
plasma levels vary significantly between individuals, and there is
no clear dose–response relationship
[31,32]
. However, high plasma
Figure 3. Relative change in mean apixaban concentration (red), albumin (blue), and IgG concentration (black) at 5, 15, 30, 60, and 90 min of CPB, CPB weaning (x) ,
30 min after CPB weaning (e), and 24 h postoperatively, calculated from the start of CPB with hemoadsorption (t0). Error bars represent standard deviations (SD).
Statistical significances are indicated as follows: ***P<0.001, **P=0.001–0.01, *P=0.01–0.05, ns =P>0.05.
Figure 2. Apixaban concentrations intraoperatively and postoperatively. N=8. Boxes extend from the 25th to 75th percentiles. The middle line represents the
median. Whiskers extend from the minimum to the maximum values. Statistical significances are indicated as follows: ***P<0.001, **P=0.001–0.01,
*P=0.01–0.05, ns =P>0.05.
Røed-Undlien et al. International Journal of Surgery (2024)
7786
concentrations of FXa inhibitors may be correlated with an
increased risk of severe bleeding complications, and a recent
American review article suggests preoperative plasma levels of
DOACs should be <30 µg/l in patients undergoing cardiac
surgery with risk factors such as renal or hepatic impair-
ment
[20,33,34]
. Being highly protein-bound, apixaban removal by
dialysis is not a good option
[35]
. In our previously published
work, we found that the reduction of apixaban concentration by
in-vitro hemoadsorption was followed by improvements in both
rotational thromboelastometry and thrombin generation para-
meters, suggesting that reducing apixaban concentration
does have an effect on the improvement of hemostasis
[17]
.
Interestingly, the two patients in our cohort who died post-
operatively were the two with the highest apixaban concentra-
tions preoperatively (Table 1). The mean CTD volume of these
two patients was 695 ml, compared to a mean of 596 ml in the
rest of the patients, which might suggest that their bleeding ten-
dency may have been higher than the rest of the group in total.
However, when comparing the individuals, there were several
patients with CTD volumes equal to these two (Table 1).
Considering the limited sample size, we cannot further elaborate
on whether various preoperative apixaban concentrations affec-
ted the outcomes of this study, but it should be noted as a
potential confounding factor.
Apart from drug levels, numerous factors should be considered
when evaluating the potential impact of hemoadsorption.
Postoperative chest tube drainage volume has been found to be an
independent risk factor for postoperative death and other adverse
outcomes following cardiac surgery
[36]
. A high chest tube output
indicates a large blood loss into the chest cavity, which com-
promises cardiac and respiratory function
[37]
. There is one pro-
spective non-randomized study demonstrating lower chest tube
drainage (CTD) volumes in apixaban-treated patients under-
going emergency cardiac surgery using hemoadsorption, but this
study did not include aortic disease, and apixaban concentrations
were not assessed
[38]
. In our present study, we found a mean 24-h
CTD volume of 621 ±136 ml, but lacking a control group, our
study was not designed to assess CTD volumes. The newly pub-
lished initial report from the international STAR (safe and timely
antithrombotic removal) registry reported a mean 24-h CTD
volume of 651 ±407 ml in 51 DOAC-treated patients undergoing
cardiac surgery with intraoperative hemoadsorption
[39]
. Two
(3.9%) of these patients suffered from ATAAD. ATAAD triggers
a coagulopathy marked by consumption, particularly note-
worthy when combined with intraoperative deep hypothermia.
This distinctive coagulative state suggests that comparing bleed-
ing volumes in this group with other cardiac surgery patients may
not be relevant
[40]
. A retrospective study conducted in our
department and published in 2020 reported a mean CTD volume
of 820 ml in five DOAC-treated patients following surgery for
acute aortic disease
[5]
. Hemoadsorption was not applied in these
cases. However, given the limited sample size in both studies, it is
prudent to exercise caution when comparing these figures.
Three(37.5%)ofthepatientsin our cohort displayed clinical
symptoms of neurologic impairment postoperatively, and all were
confirmed as strokes with CT or MRI (Table 1). One of these
presented with neurological symptoms before surgery due to the
aortic dissection causing an occlusion of the common carotid
artery. Of the two other patients with postoperative strokes, one
had temporary symptoms and showed no signs of neurologic
impairment 3 months postoperatively. This means that one patient
(12.5%) in our cohort suffered from permanent neurologic damage
that was not present before surgery and must be considered sur-
gery-related. These findings resemble those of the aforementioned
studybyHassanet al., which examined the efficacy of hemoad-
sorption in removing ticagrelor and rivaroxaban in patients
undergoing acute surgery for ATAAD. The authors reported that
9.5% of the patients experienced persistent neurological dysfunc-
tion; however, it was not clarified whether these patients belonged
to the hemoadsorption group or the control group
[14]
. Also, it is
important to acknowledge that all eight patients in our current
study were receiving apixaban treatment for atrial fibrillation,
thereby increasing their susceptibility to stroke. Compared with our
department’s prior data, which indicated an 8.4% incidence of
postoperative stroke in 107 ATAAD patients and 25% among
warfarin-treated patients, we did not observe a disproportionately
high stroke rate in our current study
[5]
. There are no studies indi-
cating that the use of CytoSorb increases the risk of thromboem-
bolism or stroke. A randomized controlled trial from 2023, using
CytoSorb for cytokine adsorption in thoracoabdominal aortic
repair, reported no device-related adverse events (n =10 in the
intervention group)
[41]
.
A recent review article suggests 4-factor prothrombin complex
concentrate (PCC) or activated prothrombin complex con-
centrate (aPCC) as management options to reduce perioperative
bleeding in FXa inhibitor-treated patients undergoing acute car-
diovascular surgery
[42]
. A recent Cochrane review article con-
cluded that PCC “may reduce red blood cell transfusion rates”
when compared with rFVIIa in patients experiencing coagulo-
pathic, non-surgical bleeding following cardiac surgery, but none
of the studies included oral FXa inhibitors
[43]
. While larger,
randomized studies are lacking in this patient population, several
smaller studies provide evidence in favor of employing PCC/
aPCC in this particular disease context
[44,45]
. The risk of
thromboembolic complications following administration of
aPCC in cardiac surgery-related refractory bleeding has been
reported up to 4.6%
[46]
. In our cohort, all patients received PCC
or aPCC in varying dosages (Table 1). One patient received rFVII
in combination with PCC, and two patients received both PCC
and aPCC. Our study was not designed to compare the effects of
different coagulation factor concentrates, and we cannot draw
any conclusions in terms of possible adverse effects. However, it is
worth noticing that the one patient who died from fatal cere-
brovascular damage received an initial dosage of 3000 IU (24 IU/
kg) of aPCC before CPB, followed by a second dosage of 3000 IU
of aPCC due to persistent bleeding after CPB weaning. The other
patient who suffered permanent neurological impairment post-
operatively received both PCC (1200 IU, 15 IU/kg) and aPCC
(3900 IU, 50 IU/kg). This patient, however, had displayed neu-
rological symptoms preoperatively. Given that patients treated
with apixaban are already at an elevated risk of thromboembo-
lism, it is prudent to consider utilizing the lowest effective dosage
to mitigate the potential for thromboembolic complications
associated with coagulation factor concentrates. As previously
mentioned, there is a need for larger, randomized studies com-
paring different strategies for the reversal of FXa inhibitor effect
in patients undergoing cardiac surgery.
The antidote targeting FXa inhibitors, andexanet alfa, has
not been found to be easily compatible with the use of CPB, as
it promotes heparin resistance
[47]
. Studies have found inci-
dences of thromboembolic complications following antidote
administration from 5% to 10%, and though proven very
Røed-Undlien et al. International Journal of Surgery (2024)
7787
effective in reversing the anticoagulant effect of FXa inhibitors,
some cost-effectiveness analyses have concluded that
andexanet alfa does not offer significant value when compared
with standard use of PCC
[48,49]
. Additionally, when compared
to hemoadsorptive devices like CytoSorb, andexanet alfa is
substantially more expensive, with an average cost of 26 787
USD in 2023, compared to 1400 USD for a CytoSorb
cartridge
[50,51]
. Although probably linked to other pharma-
cokinetic and -dynamic mechanisms than the postoperative
rise in apixaban levels found in our present study, a gradual
increase in apixaban levels over several hours is also reported
after the administration of andexanet alfa
[30]
.
Given the coagulopathic state of an acute aortic disease, which
is aggravated by extracorporeal circulation, hemodilution, and
hypothermia, the fibrinolytic system is activated, making
administering additional fibrinogen reasonable. A recent study
found that the administration of fibrinogen concentrates in
patients with plasma fibrinogen levels <1.5 g/l undergoing aortic
arch surgery significantly reduced transfusions and short-term
CTD volumes when compared to those not receiving fibrinogen
concentrate. There was no randomization between the interven-
tion and the control group, and patients on anticoagulation
therapy were not included
[52]
. The European Association of
Cardiothoracic Anesthesiology published guidelines in 2019
recommending fibrinogen replacement therapy to maintain
physiologic levels of plasma fibrinogen, yet advocating for the
avoidance of too-high plasma levels of fibrinogen
[53]
. In a Chinese
study from 2017, the authors found that low fibrinogen levels at
hospital admission were related to increased in-hospital mortality
in ATAAD patients
[54]
. It has been suggested that a longer time
from symptom onset to surgery is related to increased coagulo-
pathy, as the activation of the coagulation cascade and fibrino-
lysis system seems to increase with the duration of the dissection.
Hence, postponing surgery to await drug elimination could also
contribute to worse outcomes
[55]
. If subsequent studies corro-
borate these findings, it may complicate the current guideline
recommendations advocating for the delay of surgery to allow for
the spontaneous clearance of oral anticoagulants
[9]
.
Patients with an increased risk of ATAAD may have atrial
fibrillation, and the indication for the use of an anticoagulant.
The anticoagulant effect of warfarin is easy to assess and reverse
when in need of emergency aortic surgery, but warfarin use comes
with the need for monitoring and several interactions with foods
and other drugs
[56]
. In patients with risk factors for ATAAD, such
as enlargement of the ascending aorta, bicuspid aortic valve, or
family history of aortic disease, the choice of oral anticoagulant
may be difficult
[57]
. Thus, effective strategies for the reversal of
FXa inhibitors in acute cardiac surgery are needed. In our present
study, we have demonstrated a reduction of plasma concentra-
tions of apixaban using intraoperative hemoadsorption by
CytoSorb, suggesting that the hemoadsorption lowers apixaban
concentrations. The hemoadsorptive cartridge can be easily
connected to the CPB circuit during surgery by a perfusionist, and
as previously noted, its cost is reasonable. Currently, practice
often involves delaying surgery to allow for drug washout, which
results in a significant mortality risk for ATAAD patients
awaiting intervention. If larger, randomized studies demonstrate
a notable impact of intraoperative hemoadsorption on clinical or
economic outcomes –such as postoperative survival rates,
transfusion volumes, and ICU stays –it would be sensible to
implement a protocol for intraoperative hemoadsorption in all
FXa inhibitor-treated patients requiring emergency cardiac sur-
gery. Future research should explore whether this protocol
should also incorporate the use of aPCC. In 2019, Gleason
et al.
[58]
published a study where two CytoSorb cartridges were
connected simultaneously in a CPB circuit, hypothetically dou-
bling the adsorption from 300 ml/min to 600 ml/min. It remains
to be seen whether this approach could enhance FXa inhibitor
removal rates in emergency cardiac surgery and, perhaps, further
decrease perioperative bleeding and the need for additional coa-
gulation factor concentrates in this clinical setting.
There are several limitations to our current study. Given the
restricted number of participants and no control group, this study
lacks the power to draw conclusions regarding the impact of
hemoadsorption during CPB, especially on clinical and financial
outcomes. Furthermore, the influence of varying usage of coa-
gulation factor concentrates on our study results remains uncer-
tain. Larger, randomized controlled trials are necessary to
address these limitations. As previously mentioned, study-specific
blood samples were collected in citrated tubes. In our results,
albumin is presented as a substitute indicator for hemodilution.
Hematocrit is conventionally the preferred measure for assessing
hemodilution accurately, and was measured throughout the
surgery following the department’s local routines. Therefore,
hematocrit was not assessed at the exact same time points as
apixaban concentration, albumin, and IgG. However, as albumin
and IgG demonstrated similar profiles as hematocrit, albumin
may be considered an adequate alternative for the assessment of
hemodilution in plasma samples.
In this study, the timing of the last apixaban intake was not
recorded. Considering that apixaban reaches its maximum con-
centration (C
max
) ~4 h after oral administration, and given the
shape of the concentration curve, the timing of the last intake
could potentially impact the interpretation of our results
[59]
.
Nevertheless, the repeated measurements of apixaban con-
centration using UPLC-MS throughout the surgery and post-
operatively represent a strength of our study.
In conclusion, this observational study has demonstrated a
reduction of plasma concentrations of apixaban during acute
thoracic aortic repair for ATAAD with the use of CytoSorb. As
the number of patients receiving FXa inhibitors continues to rise,
there is a need for randomized clinical trials to assess the effec-
tiveness of reversal strategies for FXa inhibitors in acute cardiac
surgery, aiming to reduce perioperative mortality and bleeding
complications. Intraoperative hemoadsorption is relatively low-
cost, can be easily integrated into a CPB circuit, and may prove to
be an effective option in this context.
Ethical approval
The study received approval from The Norwegian National
Research Ethics Committee (REK #204588). The study was
recommended by the data protection officer at Oslo University
Hospital (#20/27731).
Consent
Three participants were included based on presumed consent
following postoperative death or cognitive impairment. All other
study participants signed a written informed consent.
Røed-Undlien et al. International Journal of Surgery (2024)
7788
Source of funding
The work was supported by Oslo University Hospital and the
University of Oslo.
Author contribution
H.R.-U.: study design, data collection, data analysis, data inter-
pretation, and writing the paper; N.H.S.: study design, data
collection, data interpretation, resources, supervision, and read-
ing and reviewing the paper; I.M.H.: study design, data collec-
tion, and reading and reviewing the paper.
Birgit M. Wollmann: data analysis, data interpretation, read-
ing and reviewing the paper; R.R.A.: data analysis and statistics,
supervision, data interpretation, and reading and reviewing the
paper; E.M.: study design, resources, data interpretation, and
reading and reviewing the paper; E.K.A.: resources, data inter-
pretation, and reading and reviewing the paper; J.L.B.: study
design, data interpretation, supervision, resources, and reading
and reviewing the paper.
Conflicts of interest disclosure
The authors declare no conflicts of interest.
Research registration unique identifying number
(UIN)
Clinical trial registration number (www.clinicaltrials.gov):
NCT05235178.
Guarantor
Johannes Lagethon Bjørnstad and Henriette Røed-Undlien.
Data availability statement
Anonymous data that support the findings of this study may be
made available upon reasonable request to the corresponding
author. The data are not publicly available due to privacy or
ethical restrictions.
Provenance and peer review
Not commissioned, externally peer-reviewed.
Acknowledgements
The authors would like to thank the perfusionists, surgeons,
anesthesiologists, and nurses at the Department of
Cardiothoracic Surgery, Oslo University Hospital, for their
valuable support and contribution to this study.
References
[1] Fang MC, Reynolds K, Fan D, et al. Clinical outcomes of direct oral
anticoagulants vs warfarin for extended treatment of venous throm-
boembolism. JAMA Network Open 2023;6:e2328033.
[2] Granger CB, Alexander JH, McMurray JJV, et al. Apixaban versus warfarin
in patients with atrial fibrillation. N Engl J Med 2011;365:981–2.
[3] Vinogradova Y, Coupland C, Hill T, et al. Risks and benefits of direct oral
anticoagulants versus warfarin in a real world setting: cohort study in
primary care. BMJ 2018;362:k2505.
[4] Bakken GV, Berg CL, Blix HS, et al. Drug Consumption in Norway
2017–2021 - data from Norwegian Drug Wholesales Statistics and the
Norwegian Prescription Database. Norwegian Institute of Public Health;
2022. (978-82-8406-313-3).
[5] Bjørnstad JL, Khan AM, Røed-Undlien H, et al. Operative survival in
patients with acute aortic disease in the era of newer oral anticoagulants.
Open Heart 2020;7:e001278.
[6] Sromicki J, Van Hemelrijck M, Schmiady MO, et al. Prior intake of new
oral anticoagulants adversely affects outcome following surgery for acute
type A aortic dissection. Interact Cardiovasc and Thorac Surg 2022;35:
ivac037.
[7] Thalji NK, Patel PA, Camire RM. Administration of andexanet alfa for
direct factor Xa inhibitor reversal precludes therapeutic heparinization
for cardiopulmonary bypass. Blood 2019;134(Suppl_1):3640.
[8] Watson CJ, Zettervall SL, Hall MM, et al.Difficult intraoperative
heparinization following andexanet alfa administration. Clin Pract Cases
Emerg Med 2019;3:390–4.
[9] Steffel J, Collins R, Antz M, et al. 2021 European Heart Rhythm
Association Practical Guide on the use of non-vitamin K antagonist oral
anticoagulants in patients with atrial fibrillation. Europace 2021;23:
1612–76.
[10] Harris KM, Nienaber CA, Peterson MD, et al. Early mortality in type A
acute aortic dissection: insights from the international registry of acute
aortic dissection. JAMA Cardiol 2022;7:1009–5.
[11] Obel LM, Lindholt JS, Lasota AN, et al. Clinical characteristics, inci-
dences, and mortality rates for type A and B aortic dissections: a
nationwide Danish population-based cohort study from 1996 to 2016.
Circulation 2022;146:1903–7.
[12] Liu Y, Han L, Li J, et al. Consumption coagulopathy in acute aortic
dissection: principles of management. J Cardiothorac Surg 2017;12:50.
[13] Hassan K, Bayer N, Schlingloff F, et al. Bleeding complications after use
of novel oral anticoagulants in patients undergoing cardiac surgery. Ann
Thorac Surg 2018;105:702–8.
[14] Hassan K, Brüning T, Caspary M, et al. Hemoadsorption of rivaroxaban
and ticagrelor during acute type A aortic dissection operations. Ann
Thorac Cardiovasc Surg 2022;28:186–92.
[15] Hassan K, Kannmacher J, Wohlmuth P, et al. Cytosorb adsorption
during emergency cardiac operations in patients at high risk of bleeding.
Ann Thorac Surg 2019;108:45–51.
[16] Krüger B, Renner T, Van Hemelrijck M, et al. The effect of hemoad-
sorption on rivaroxaban blood plasma concentration in emergency car-
diac surgery. Indian J Thorac Cardiovasc Surg 2021;37:680–3.
[17] Røed-Undlien H, Schultz NH, Lunnan A, et al. In vitro apixaban removal by
CytoSorb whole blood adsorber: an experimental study. J Cardiothorac
Vasc Anesth 2022;36:1636–44.
[18] Buonocore M, Rex S, Degezelle K, et al. CytoSorb haemoadsorption for
removal of apixaban—a proof-of-concept pilot case for a randomized
controlled trial. J Clin Pharm Ther 2022;47:2373–5.
[19] Nardi M, Damato I, Caso A, et al. Apixaban blood levels reduced by the
use of Cytosorb added to cardiopulmonary bypass. Blood Purif 2022;51:
2–100; abstract A67.
[20] Moster M, Bolliger D. Perioperative guidelines on antiplatelet and antic-
oagulant agents: 2022 update. Curr Anesthesiol Rep 2022;12:286–96.
[21] Levy JH, Ageno W, Chan NC, et al. When and how to use antidotes for
the reversal of direct oral anticoagulants: guidance from the SSC of the
ISTH. J Thromb Haemost 2016;14:623–7.
[22] Mathew G, Agha R, Albrecht J, et al. STROCSS 2021: Strengthening the
reporting of cohort, cross-sectional and case–control studies in surgery.
Int J Surg 2021;96:106165.
[23] Setup of CytoSorb: CytoSorbents Europe GmbH 2024. https://cytosorb-
therapy.com/en/the-adsorber/setup-of-cytosorb/
[24] DeBakey ME, McCollum CH, Crawford ES, et al. Dissection and dis-
secting aneurysms of the aorta: twenty-year follow-up of five hundred
twenty-seven patients treated surgically. Surgery 1982;92:1118–34.
[25] Lin C-Y, Tung T-H, Wu M-Y, et al. Surgical outcomes of DeBakey type I
and type II acute aortic dissection: a propensity score-matched analysis in
599 patients. J Cardiothorac Surg 2021;16:208.
[26] Hickey M, Gatien M, Taljaard M, et al. Outcomes of urgent warfarin
reversal with frozen plasma versus prothrombin complex concentrate in
the emergency department. Circulation 2013;128:360–4.
[27] Karkouti K, Beattie WS, Wijeysundera DN, et al. Hemodilution during
cardiopulmonary bypass is an independent risk factor for acute renal
Røed-Undlien et al. International Journal of Surgery (2024)
7789
failure in adult cardiac surgery. J Thorac Cardiovasc Surg 2005;129:
391–400.
[28] Ranucci M, Baryshnikova E, Ciotti E, et al. Hemodilution on cardio-
pulmonary bypass: thromboelastography patterns and coagulation-
related outcomes. J Cardiothorac Vasc Anesth 2017;31:1588–94.
[29] Strunden MS, Heckel K, Goetz AE, et al. Perioperative fluid and volume
management: physiological basis, tools and strategies. Ann Intensive
Care 2011;1:2.
[30] Connolly SJ, Crowther M, Eikelboom JW, et al. Full study report of
andexanet alfa for bleeding associated with factor Xa inhibitors. N Engl J
Med 2019;380:1326–35.
[31] Mavri A, Vene N, Božič-Mijovski M, et al. Apixaban concentration
variability and relation to clinical outcomes in real-life patients with atrial
fibrillation. Sci Rep 2021;11:13908.
[32] Helin T, Joutsi-Korhonen L, Lassila R. Clinical use and laboratory testing of
oral anticoagulation therapy: experience from Finland. Ann Blood 2019;4:17.
[33] Suwa M, Morii I, Kino M. Rivaroxaban or apixaban for non-valvular
atrial fibrillation - efficacy and safety of off-label under-dosing according
to plasma concentration. Circ J 2019;83:991–9.
[34] Leil TA, Frost C, Wang X, et al. Model-based exposure-response analysis of
apixaban to quantify bleeding risk in special populations of subjects undergoing
orthopedic surgery. CPT Pharmacometrics Syst Pharmacol 2014;3:e136.
[35] Lidgard B, Shen JI. Apixaban dosing in patients with kidney failure
treated with peritoneal dialysis. Kidney Med 2023;5:100699.
[36] Christensen MC, Dziewior F, Kempel A, et al. Increased chest tube
drainage is independently associated with adverse outcome after cardiac
surgery. J Cardiothorac Vasc Anesth 2012;26:46–51.
[37] Dixon B, Santamaria JD, Reid D, et al. The association of blood trans-
fusion with mortality after cardiac surgery: cause or confounding?
(CME). Transfusion 2013;53:19–27.
[38] Hassan K, Thielmann M, Easo J, et al. Removal of apixaban during
emergency cardiac surgery using hemoadsorption with a porous polymer
bead sorbent. J Clin Med 2022;11:5889.
[39] Schmoeckel M, Thielmann M, Hassan K, et al. Intraoperative hae-
moadsorption for antithrombotic drug removal during cardiac surgery:
initial report of the international safe and timely antithrombotic removal
(STAR) registry. J Thromb Thrombolysis 2024;57:888–97.
[40] Zindovic I, Sjögren J, Bjursten H, et al. The coagulopathy of acute type A
aortic dissection: a prospective, observational study. J Cardiothorac Vasc
Anesth 2019;33:2746–54.
[41] Doukas P, Hellfritsch G, Wendt D, et al. Intraoperative hemoadsorption
(Cytosorb™) during open thoracoabdominal aortic repair: a pilot ran-
domized controlled trial. J Clin Med 2023;12:546.
[42] Heuts S, Ceulemans A, Kuiper G, et al. Optimal management of cardiac
surgery patients using direct oral anticoagulants: recommendations for
clinical practice. Eur J Cardiothorac Surg 2023;64:ezad340.
[43] Hayes K, Fernando MC, Jordan V. Prothrombin complex concentrate in
cardiac surgery for the treatment of coagulopathic bleeding. Cochrane
Database Syst Rev 2022;11:CD013551.
[44] Song Y, Wang Z, Perlstein I, et al. Reversal of apixaban anticoagulation by
four-factor prothrombin complex concentrates in healthy subjects: a rando-
mized three-period crossover study. J Thromb Haemost 2017;15:2125–37.
[45] Schultz NH, Tran HTT, Bjørnsen S, et al. The reversal effect of prothrombin
complex concentrate (PCC), activated PCC and recombinant activated
factor VII against anticoagulation of Xa inhibitor. Thromb J 2017;15:6.
[46] Khoury W, Servito M, Wang L, et al. The use of FEIBA for refractory
bleeding in cardiac surgery - a systematic review. Expert Rev Cardiovasc
Ther 2022;20:403–8.
[47] Eche IM, Elsamadisi P, Wex N, et al. Intraoperative unfractionated
heparin unresponsiveness during endovascular repair of a ruptured
abdominal aortic aneurysm following administration of andexanet alfa
for the reversal of rivaroxaban. Pharmacotherapy 2019;39:861–5.
[48] Frontera JA, Bhatt P, Lalchan R, et al. Cost comparison of andexanet
versus prothrombin complex concentrates for direct factor Xa inhibitor
reversal after hemorrhage. J Thromb Thrombolysis 2020;49:121–31.
[49] Micieli A, Demchuk AM, Wijeysundera HC. Economic evaluation of
andexanet versus prothrombin complex concentrate for reversal of factor
Xa-associated intracranial hemorrhage. Stroke 2021;52:1390–7.
[50] Rao C, Preissing F, Thielmann M, et al. Hemoadsorption using CytoSorb
(®) in patients with infective endocarditis: a German-based budget
impact analysis. J Cardiovasc Dev Dis 2023;10:366.
[51] Andexanet Alfa (Ondexxya). CADTH Reimbursement Recommendation.
Canadian Agency for Drugs and Technologies in Health; 2023 https://
www.ncbi.nlm.nih.gov/books/NBK599838/
[52] Guan X, Li L, Lu X, et al. Safety and efficacy of fibrinogen concentrate in
aortic arch surgery involving moderate hypothermic circulatory arrest. J
Thromb Thrombolysis 2023;55:67–73.
[53] Erdoes G, Koster A, Meesters MI, et al. The role of fibrinogen and
fibrinogen concentrate in cardiac surgery: an international consensus
statement from the Haemostasis and Transfusion Scientific
Subcommittee of the European Association of Cardiothoracic
Anaesthesiology. Anaesthesia 2019;74:1589–600.
[54] Liu J, Sun L-L, Wang J, et al. The relationship between fibrinogen and in-
hospital mortality in patients with type A acute aortic dissection. Am J
Emerg Med 2018;36:741–4.
[55] Zindovic I, Sjögren J, Bjursten H, et al. Predictors and impact of massive
bleeding in acute type A aortic dissection. Interact Cardiovasc Thorac
Surg 2017;24:498–505.
[56] Wells PS, Holbrook AM, Crowther NR, et al. Interactions of warfarin
with drugs and food. Ann Intern Med 1994;121:676–83.
[57] Czerny M, Grabenwöger M, Berger T, et al. EACTS/STS Guidelines for
diagnosing and treating acute and chronic syndromes of the aortic organ.
Eur J Cardiothorac Surg 2024;65:ezad426.
[58] Gleason TG, Argenziano M, Bavaria JE, et al. Hemoadsorption to reduce
plasma-free hemoglobin during cardiac surgery: results of REFRESH I
pilot study. Semin Thorac Cardiovasc Surg 2019;31:783–93.
[59] Byon W, Garonzik S, Boyd RA, et al. Apixaban: a clinical pharmacokinetic
and pharmacodynamic review. Clin Pharmacokinet 2019;58:1265–79.
Røed-Undlien et al. International Journal of Surgery (2024)
7790
