REVIEW
A Review of Topical Hemostatic Agents for Use in
Cardiac Surgery
James Barnard, MD, MRCS, and Russell Millner, MD, FRCS(CTh)
Department of Cardiothoracic Surgery, Lancashire Cardiac Centre, Blackpool Victoria Hospital, Blackpool, United Kingdom
Postoperative hemorrhage, redo sternotomy for bleeding,
and transfusion of blood products are all associated with
poorer outcomes in cardiac surgery. Topical hemostatic
agents are important adjuncts to reduce blood loss after
cardiac surgery and can have a role in reducing both
“surgical” and “nonsurgical bleeding.” There are many
H emostatic agents are a useful tool in reducing peri-
operative and postoperative bleeding. Patients
coming forward for urgent cardiac surgery may have
been exposed to aspirin, clopidogrel, or glycoprotein
IIb–IIIa inhibitors, in addition to heparin or warfarin
resulting in a higher degree of postoperative bleeding [1].
Prolonged time on cardiopulmonary bypass, cooling, and
heparinization all contribute to the derangement of blood
clotting mechanisms and predispose the patient to post-
operative bleeding. Excessive bleeding may necessitate
transfusion of blood products with potential associated
adverse effects, including anaphylactic transfusion reac-
tions, transfusion-related lung injury, and transmission
of infectious agents. Postoperative hemorrhage, transfu-
sion of blood products, and re-exploration for bleeding
[2] are all associated with poorer outcome in cardiac
surgery. Surgical causes of postoperative bleeding are
found in 66% of cases on re-exploration with the remain-
der being attributed to coagulopathy [3]. Surgical bleed-
ing is considered to be bleeding at the site of a surgical
suture line or anastomosis, whereas nonsurgical bleeding
is hemorrhaging of small vessels from a tissue surface
and is generally accepted as being as a result of a
coagulopathy. A variety of topical hemostatic agents are
used with varying efficacy and varying degrees of evi-
dence in favor of their benefits (Table 1). Several new
agents have been approved for use in the last few years,
including chitosan-based agents, which have shown
great promise in controlling major hemorrhaging in the
pre-hospital setting and in animal models of major hem-
orrhaging [4]. The aim of this review is to discuss agents
that are presently used and agents that are also presently
used in other fields and may have an important role to
play in cardiac surgery.
Address correspondence to Mr Millner, Department of Cardiothoracic
Surgery, Blackpool Victoria Hospital, Whinney Heys Rd, Blackpool, FY3
8NR, United Kingdom; e-mail: russellmillner@btinternet.com.
© 2009 by The Society of Thoracic Surgeons
Published by Elsevier Inc
topical hemostatic agents to choose from, and with sev-
eral new products in this field being approved for use in
the last few years, the aim of this review is to appraise
these agents and to look at the evidence for their efficacy.
(Ann Thorac Surg 2009;88:1377– 83)
© 2009 by The Society of Thoracic Surgeons
Methods
A systematic English language literature search was con-
ducted using MEDLINE and PubMed between January
1974 and August 2008. Keywords used for the search in-
cluded bleeding, hemostatic agent, sealant, and cardiac
surgery. Additional reports were identified by references
cited in the publications found using the search terms and
also in published review articles. Emphasis was placed on
randomized controlled studies where possible.
Compression Hemostatic Agents
Compression hemostatic agents absorb blood and pro-
vide a scaffold for clot formation in addition to exerting
compression on a bleeding site.
Oxidized Regenerated Cellulose
Surgicel, Oxycel
REVIEW
Surgicel (Johnson & Johnson Medical Inc, Arlington, TX)
is a local hemostatic bioabsorbable gauze consisting of
oxidized regenerated cellulose. It is a poly anion, the
functional unit of which is polyanhydroglucuronic acid.
Surgicel provides a scaffold on which platelets can com-
mence the adhesion and aggregation process leading to
coagulation, but it is not reported to alter the physiolog-
ical clotting mechanism. There is evidence of absorption
within the first day, and complete absorption can be
observed between 4 to 8 weeks, the rate being dependent
on the volume of surgicel, local vascularity, and the
nature of the tissue bed. Surgicel is also reported to be
bacteriostatic [5] with reported antimicrobial activity,
even against antibiotic-resistant micro-organisms includ-
ing MRSA [6].
Mr Millner discloses that he has a financial relation-
ship with MedTrade Products Limited.
0003-4975/09/$36.00
doi:10.1016/j.athoracsur.2009.02.092
 1378 REVIEW BARNARD AND MILLNER Ann Thorac Surg
TOPICAL HEMOSTATIC AGENTS FOR USE IN CARDIAC SURGERY 2009;88:1377– 83

Table 1. Summary of Randomized Controlled Trials of Hemostatic Agents in Cardiovascular Surgery

Agent Control Number Outcome

Colgel [7] Surgicel 71a

Chest tube drainage in first 24 hr was Colgel 373 ⫾ 143 mL vs Surgicel 571
⫾ 144 mL, (p ⫽ 0.01) and in the first 3 hr it was 132 ⫾ 41 vs 228 ⫾ 57
mL, respectively (p ⬍ 0.001).
CoStasis [9]* Collagen sponges 72a Bleeding controlled within 10 minutes with CoStasis 28/37 vs control 17/37;
p ⫽ 0.02.
Cyanoacrylate Oxidized Cellulose 151b Hemostasis mean time 119.3 vs 403.8 seconds (p ⬍ 0.001) and immediate
Sealant [12]* hemostasis in 54.5% vs 10% in Cyanoacrylate Sealant and control,
respectively.
Tisseel [44]* Avitene/Surgicel 164a 92.6% bleeding control with Tisseel vs 12.4% with control agents at 5
minutes (p ⬍ 0.001).
Beriplast [22]* Surgery alone 52c Fewer transfusions of FFP in the BP group compared with controls (p ⱕ
0.05). BP group shorter time to achieve hemostasis (p ⱕ 0.05), less
bleeding intraoperatively (p ⱕ 0.01).
CoSeal [24]* Gelfoam thrombin 54d Immediate hemostasis in CoSeal 48 of 59 vs control 10 of 27 (p ⫽ 0.002).
BioGlue [27]* Surgery alone 151e BioGlue bleeding (18.8% of anastomoses) vs control (42.9% of anastomoses;
p ⬍ 0.001).
FloSeal [35]* Gelfoam thrombin 93a FloSeal stopped bleeding in 94% of the patients (first bleeding site only)
within 10 minutes compared with 60% in the control group (p ⫽ 0.001).
rThrombin [42]* bThrombin 164b Hemostasis at 10 minutes 94% bThrombin, 91% rThrombin; p ⫽ NS.
a b
Cardiovascular procedures including re-do cardiac operations and ascending aortic aneurysm repair. Peripheral vascular and arteriovenous
c d
procedures. Procedures for congenital heart disease. Implantation of Dacron grafts (DuPont, Wilmington, DE) for the repair of nonruptured
e
aneurysms. Cardiac and vascular repair procedures.

* Studies sponsored by industry.

BP ⫽ Beriplast fibrin sealant; bThrombin ⫽ bovine plasma-derived thrombin; FFP ⫽ fresh frozen plasma; NS ⫽ not statistically significant;
rThrombin ⫽ recombinant thrombin.

Collagen thrombin in a calcium chloride buffer mixed intraop-

eratively with an equal volume of autologous plasma
Colgel, Helitene, Avitene
obtained at the time of operation. The patient’s plasma
Microfibrillar collagen was initially described in 1969.
provides fibrinogen that is cleaved by the thrombin to
It is a water-insoluble acid salt of bovine collagen,
form a collagen-fibrin gel matrix. The composite is
which when used topically adheres to the bleeding site
delivered as a spray onto the bleeding site. CoStasis
and provides some hemostatic effect and initiates
(Cohesion Technologies) has been compared with the
platelet activation and aggregation, as well as reinforc-
use of gauze and collagen sponges and has been shown
ing the fibrin clot that is formed. For high transfusion
to achieve hemostasis within 3 minutes in cardiac
risk cardiac surgery patients, including repeat cardiac
REVIEW

surgery patients undergoing cardiopulmonary bypass

operations (such as aorta coronary bypass operations
or valvular operations), ascending aortic aneurysm
repair necessitating deep hypothermic circulatory ar-
rest and ascending aortic grafting without deep hypo-
thermic circulatory arrest Surgicel has been compared
with microfibrillar collagen hemostat (Colgel [Labora-
torie Interphar, Aubervilliers, France]) in terms of the
ability to reduce postoperative bleeding and was found
to be superior in terms of reducing postoperative
drainage (Table 1) [7]. Microfibrillar collagen is not as
good at reducing infection risk as Surgicel [5] and is
reported to cause end-organ damage if shed blood
containing it is returned to the circulation by either
pump suction or cell salvage devices, as it may readily
pass through the filters of these devices [8].
Composite Collagen and Thrombin Sealant
CoStasis
CoStasis (Cohesion Technologies, Palo Alto, CA) is a
composite of bovine microfibrillar collagen and bovine
for coronary artery bypass grafting in 76% of cases
compared with 46% achieved in the control group [9].
The study does not give exact details regarding the site
or timing of application, and the use of CoStasis did
not have any impact on reducing transfusion of blood
products.
Thrombin Sealants
Thrombin, Thrombogen, Thrombistat
It is estimated that thrombin has been used as an
adjunctive hemostat in more than half a million surgi-
cal procedures per year in the United States [10].
Thrombin is a serum protease that transforms fibrino-
gen into fibrin, and activates factor XIII and promotes
stabilization of the clot formed by fibrin and other
proteins. The majority of thrombin used is derived
from bovine plasma, although in January 2008, recom-
binant human thrombin (Recothrom [ZymoGenetics,
Seattle, WA]) gained United States Food and Drug
Ann Thorac Surg
2009;88:1377– 83 TOPICAL HEMOSTATIC AGENTS FOR USE IN CARDIAC SURGERY
Administration (FDA) approval and has potential ad-
vantages in comparison with the bovine or human
types, including reduction in the formation of patient
thrombin antibodies and exclusion of the risk of trans-
mission of infection while maintaining comparable
efficacy [11].
Cyanoacrylate Sealants
Omnex
Omnex (Ethicon Surgical Sealant; Ethicon Inc, Somer-
ville, NJ) is a synthetic tissue adhesive consisting of a
blend of two monomers (ie, 2-octyl cyanoacrylate and
butyl lactoyl cyanoacrylate). The liquid preparation is
contained inside a crushable capsule that delivers the
preparation through a porous disc, which contains an
initiator agent to activate the delivered sealant. The
sealant polymerizes to form a flexible sealing film, which
is adherent to both synthetic material and human tissue
in a process that is independent of the patient’s clotting
processes. The seal degrades with time, breaking down
into smaller absorbable fragments.
A prospective, randomized controlled trial has as-
sessed the efficacy of cyano-acrylate sealants in establish-
ing hemostasis of expanded polytetrafluoroethylene to
arterial vascular anastomoses in arteriovenous grafts and
femoral bypass grafts in 151 patients [12]. The mean time
from clamp release to hemostasis was 119.3 seconds with
cyanoacrylate surgical sealant versus 403.8 seconds, with
the control (oxidized cellulose) (p ⬍ 0.001). Immediate
hemostasis was achieved in 54.5% of patients receiving
cyanoacrylate surgical sealant and in 10% of those receiv-
ing the control. A subsequent study in patients undergo-
ing arteriovenous access or vascular reconstruction be-
low the diaphragm with the use of cyanacrylate sealant
found that immediate hemostasis was achieved in 71.3%
(112 of 157) of sites and in 93.6%, 96.8%, and 100% within
1, 5, and 10 minutes, respectively [13].
Fibrin Sealants
Tisseel, Beriplast, Hemaseel, Crosseal/Quixil, Vivostat
Fibrin sealant was developed in 1972 by Matras and
colleagues [14] by successfully using a fibrinogen cryo-
precipitate in peripheral nerve anastomoses on animal
models. Fibrin sealants are made of two components
contained in separate vials: (1) a freeze dried concentrate
of clotting proteins, mainly fibrinogen, factor XIII, and
fibronectin (ie, the sealant) and (2) freeze dried thrombin
(ie, the catalyst). Fibrinogen is a precursor of fibrin, which
represents the basic element of the clot. The transforma-
tion of fibrinogen into stable fibrin occurs by means of
thrombin and factor XIII, which in turn are activated by
thrombin. Fibrin sealants emulate the final stages of the
clotting cascade, but as they are autonomous of the
body’s clotting mechanism, they are also effective in
patients with coagulopathies or those who are receiving
heparin or anticoagulants [15, 16].
Historically there have been concerns about the use of
REVIEW BARNARD AND MILLNER 1379
fibrin sealants with so-called homemade preparations
used until the 1990s, predating the United States FDA
approval, containing cryoprecipitate not subjected to
viral inactivation and containing variable levels of impu-
rities [17]. Tisseel (Baxter Healthcare, Fremont, CA) and
Beriplast (CSL Behring, King of Prussia, PA) contain
bovine aprotinin, whereas Crosseal (Ethicon Inc), mar-
keted in the United Kingdom as Quixil, contains tranex-
amic acid. These agents are added as antifibrinolytic
agents to slow fibrinolysis of the clot, which is produced
at the site of application. After a recent report of in-
creased mortality with the use of aprotinin in cardiac
surgery patients [18], Bayer Pharmaceuticals have noti-
fied the FDA of their intent to remove their systemically
administered aprotinin agent Trasylol from hospital
pharmacies. Tisseel (Baxter Healthcare) and Beriplast
(CSL Behring) are not affected by this withdrawal, al-
though there are reports of the adverse effects of bovine
aprotinin in sealants relating to the highly immunogenic
nature of bovine aprotinin in humans. Over 125 anaphy-
lactic reactions having being reported in the literature,
with an estimated anaphylaxis incidence of between 0.5
and 5 per 100,000 topical applications [15]. The Vivostat
system (Vivostat A/S, Birkerod, Denmark) is a medical
device to produce an autologous fibrin sealant CE
marked for use in Europe in 2000. The FDA approval for
use of Vivostat (Vivostat A/S) in the United States has
been applied for but was not yet granted at the time this
article was written. The device can generate 4.5 mL of
sealant from 120 mL of the patient’s blood in 23 minutes
[19]. The sealant is applied using a spray system. There is
no randomized controlled trial using this system in
cardiac surgery in which it is compared with a compara-
ble sealing agent, although its tensile strength has been
compared with other fibrin sealants and has been found
to be comparable [20].
Kjaergard and Fairbrother [21] have reported 24 trials
of the use of fibrin glues in cardiothoracic surgery, with
REVIEW
20 trials reporting a positive benefit in terms of reducing
bleeding, 4 studies reporting no difference from controls,
and no studies reporting a negative influence on bleed-
ing. More recently a randomized controlled trial in pedi-
atric cardiac surgery cases with a proven coagulopathy
has shown a significant reduction in the use of blood
products and operating room time to achieve hemostatsis
when fibrin sealant has been used [22].
A note of caution should also be raised in the use of
fibrin sealants in coronary artery bypass grafting as
Lamm and colleagues [23] have reported an increased
risk of myocardial injury or even death in cornary artery
bypass grafting patients when Tissucol fibrin sealant was
used intraoperatively. They reported that in several cases
they had an acute occlusion of bypass grafts after Tissu-
col was applied in proximity to the anastomoses with
immediate embolectomy showing fresh fibrin clot in the
lumen of the grafts in each case. This study resulted in
the product description being changed to specifically
emphasize that special care was necessary when using
Tissucol in aortocoronary bypass operations.
 1380 REVIEW BARNARD AND MILLNER
TOPICAL HEMOSTATIC AGENTS FOR USE IN CARDIAC SURGERY
Polyethylene Glycol Polymers
CoSeal, DuraSeal
Polyethylene glycol polymers are synthetic hydrogels
intended specifically as tissue sealants (ie, derivatives of
polyethylene glycol); they can propagate rapid cross
linking with inherent proteins, such as collagen, to form
a cohesive matrix that adheres strongly to the applied
tissue. The sealing capability of the hydrogel is almost
immediate and does not require any human blood prod-
ucts or bovine components to inhibit bleeding. In a
randomized controlled trial, the effectiveness of CoSeal
(Cohesion Technologies, Palo Alto, CA) was compared
with Gelfoam (Upjohn, Kalamazoo, MI) and thrombin for
managing anastomotic bleeding after implantation of
Dacron grafts (DuPont, Wilmington, DE) during aortic
reconstruction for nonruptured aneurysms. A signifi-
cantly greater proportion of bleeding suture line sites
treated with CoSeal (Cohesion Technologies) achieved
immediate sealing after re-establishment of bloodflow
[24]. Another randomized controlled trial of CoSeal ver-
sus Gelfoam (Upjohn) and thrombin during surgical
placement of prosthetic vascular grafts showed that over-
all 10-minute sealing success was equivalent. However,
subjects treated with CoSeal achieved immediate anas-
tomotic sealing at more than twice the rate of subjects
treated with Gelfoam and thrombin [25]. Duroseal is
rarely used in cardiac surgery, although an experimental
trial was performed to examine whether it had any
benefits in reducing adhesion formation post-cardiac
surgery and no benefit was proven [26].
Bovine Albumin and Gluteraldehyde
BioGlue
BioGlue (CryoLife Inc, Atlanta, GA) is a bovine serum
albumin and glutaraldehyde tissue adhesive. Two sepa-
rate solutions are dispensed by a controlled delivery
REVIEW
system, composed of a double-chambered syringe. Once
dispensed, the adhesive solutions are mixed within the
applicator tip where cross-linking begins. The glutaral-
dehyde molecules covalently bond the bovine serum
albumin molecules to each other and, upon application,
to the tissue proteins at the repair site, creating a flexible
mechanical seal independent of the body’s clotting
mechanism. BioGlue (CryoLife Inc) begins to polymerize
within 20 to 30 seconds and reaches its bonding strength
within 2 minutes. BioGlue also adheres to synthetic graft
materials through mechanical interlocks within the inter-
stices of the graft matrix.
A randomized controlled trial has compared the use of
BioGlue with a standard control repair of anastomotic
bleeding in patients undergoing cardiac and vascular
repair. In this study, 49 patients underwent cardiac pro-
cedures, 105 aortic procedures, and 48 peripheral vascu-
lar procedures. Anastomotic bleeding was significantly
reduced in the BioGlue group (18.8% of anastomoses)
compared with the control group (42.9% of anastomoses;
p ⬍ 0.001). Pledget use was reduced in the BioGlue group
Ann Thorac Surg
2009;88:1377– 83
(26.2%) compared with the control group (35.9%; p ⫽
0.047) [27].
Adverse effects have been reported with the use of
BioGlue including nerve tissue injury [28], leaking of the
BioGlue through needle holes [29], and impaired aortic
growth causing anastomotic strictures [30]. BioGlue is not
recommended in pediatric surgery due to its inability to
allow growth of the tissue it is used on. Caution must be
exercised in the application of BioGlue as it may spread
away from the site of application and has even spread
onto valve prostheses causing intraoperative dysfunction
[31, 32].
There have been concerns raised regarding the poten-
tially harmful effects of BioGlue to local tissues. Anec-
dotal reports detail the persistence of BioGlue years after
application surrounded by unusually soft and friable
native tissues suggesting that BioGlue may impair nor-
mal healing [30, 33, 34].
Gelatin and Thrombin
FloSeal
FloSeal Hemostatic Matrix (Baxter Healthcare Corp,
Freemont, CA) is a combination of a bovine-derived
gelatin-based matrix and a human-derived thrombin
component. On contact with blood, the gelatin particles
swell and tamponade the bleeding, while the high-
thrombin levels quicken clot formation. The thrombin
used in FloSeal (Baxter Healthcare Corp) is manufac-
tured from pooled human plasma subjected to a two-step
vapor-heat treatment that has been demonstrated to
significantly reduce viral load; however, as stated by the
manufacturer, no procedure has been shown to be com-
pletely effective in removing viral infectivity from deriv-
atives of human plasma. FloSeal is completely absorbed
within 6 to 8 weeks of application.
In a prospective, randomized study, conducted by Oz
and colleagues [35], 93 patients undergoing cardiac op-
erations were randomized into the FloSeal or control
group after standard surgical means to control bleeding
had failed. FloSeal stopped bleeding in 94% of the pa-
tients (first bleeding site only) within 10 minutes, com-
pared with 60% in the control group who received
Gelfoam thrombin (p ⫽ 0.001). At 3 minutes, successful
hemostasis was achieved in 72% of the FloSeal group
compared with 23% in the control group (p ⫽ 0.0001).
Success in the FloSeal group was achieved in 26 of 27
patients treated after protamine reversal of heparin and
in 17 of 19 patients treated before heparin reversal. In the
control group, hemostasis success was achieved in 21 of
28 patients treated after protamine reversal of heparin
and in 5 of 14 patients treated before heparin reversal.
The success of hemostasis in the FloSeal group was
statistically, significantly better than the control group
before protamine reversal of heparin (p ⫽ 0.0023). Suc-
cess rates between the two groups did not reach statisti-
cal significance after protamine reversal of heparin (p ⫽
0.051) [35].
Ann Thorac Surg
2009;88:1377– 83 TOPICAL HEMOSTATIC AGENTS FOR USE IN CARDIAC SURGERY
New Agents: Microporous Polysaccharide
Hemospheres
Arista AH, HemoStase MPH
Arista AH (Medafor Inc, Minneapolis, MN) and Hemo-
Stase MPH (Cryolife Inc) are simple, safe, and effective
hemostatic powders recently approved by the FDA for
most types of surgery including cardiac, orthopedic,
spinal, and general surgical applications. Microporous
polysaccharide hemospheres are plant-based and are
delivered as a flowable powder engineered to rapidly
dehydrate blood, enhancing clotting on contact. This
osmotic action causes the particle to swell and concen-
trates serum proteins, platelets, and other formed ele-
ments on its surface. The particles and their coating of
compacted cells create scaffolding for the formation of a
tenacious fibrin clot within minutes of application. The
particles are fully absorbed and enzymatically cleared
from the wound site within 24 to 48 hours. There are no
trials reporting the use of these agents for hemostasis in
cardiac surgery published in the literature.
Chitin and Chitosan-Based Hemostatic Agents
HemCon, Closure, Chitoseal, or Celox
Chitin polymer (␤-(1¡4)-N-acetyl-D-glucosamine) is a
naturally occurring polysaccharide synthesised by a large
number of living organisims and exists as a crystalline
microfibril forming structural components in the exoskel-
eton of arthropods and in the cell walls of fungi and
yeasts. Chitosan was first reported to be of potential use
in cardiovascular surgery in 1983 when it was shown to
form a coagulum when in contact with defibrinated
blood, heparinized blood, and washed red cells. When
knitted DeBakey grafts were treated with Chitosan, they
were found to be impermeable to blood [36].
Reports have indicated that Chitin and Chitosan accel-
erate blood coagulation in vivo, and that in addition they
enhance the release of platelet derived growth factor-AB
and transforming growth factor-␤1, which play important
roles in the wound healing process [37]. Coagulation and
haemagglutination tests have shown that the hemostatic
mechanism of Chitiosan seems to be independent of the
classical coagulation cascade and seems to be an inter-
action between the cell membrane of erythrocytes and
Chitsoan [38].
Celox (SAM Medical Products, Portland, OR) is a
granular hemostatic agent that interacts directly with red
blood cells and platelets to form a cross-linked barrier
clot, independent of native factors. It is reported by the
manufacturer to be nonallergenic, nonexothermic, able
to function in a hypothermic environment, and low in
cost. HemCon (HemCon Medical Technologies Inc, Port-
land, OR) is one of the two most common hemostatic
dressings used by the United States military in treating
casualties in the pre-hospital setting, and it is coated with
Chitosan. QuikClot (Z-Medica Corp, Wallingford, CT)
contains zeolite powder that absorbs water from the
blood, thereby concentrating clotting factors. A random-
ized controlled trial using a porcine model of acute
REVIEW BARNARD AND MILLNER 1381
arterial and venous hemorrhage has compared hemor-
rhage control, rebleeding, and survival between Celox
(SAM Medical Products), HemCon (HemCon Medical
Technologies Inc), QuikClot (Z-Medica Corp), and stan-
dard gauze dressings. All agents controlled initial vascu-
lar bleeding; Celox reduced rebleeding to 0% (p ⬍ 0.001);
HemCon reduced rebleeding to 33% (p ⫽ 0.038); Quik-
Clot reduced rebleeding to 8%. Celox improved survival
to 100% compared with 67% for HemCon and 92% for
Quik Clot. At our institution we have used Celox as a
topical hemostatic agent in cardiac surgery patients and
have found it easy to use and highly efficacious [39].
Comment
Choice of one topical hemostatic agent in comparison
with another is variable from one surgeon to another and
may depend on individual experiences and availability
rather than a strong evidence basis. Efficacy in the control
of bleeding within a 10-minute timeframe has been
reported for thrombin (95%) [11], Floseal (96%) [40], and
CoStasis (97%) [9]. A direct comparison has been made
between Floseal, Gelfoam, Avitene, Surgicel, and fibrin
sealants in an animal model of aortic arterial bleeding in
the presence of heparin. In this model fibrin sealant was
proven to be the most efficacious hemostatic agent [16].
However, detailed comparisons between hemostatic
agents in animal studies are difficult, as clotting mecha-
nisms outside of primate studies differ substantially.
The published studies in this field are heavily domi-
nated by trials that have been funded by the manufac-
tures of the hemostatic agents that are being tested
(Table 1). In the absence of well-controlled clinical trials,
a sound evidence basis for the use of one agent in
comparison with another will not emerge. The Society of
Thoracic Surgeons Blood Conservation Guideline Task
Force have recommended that topical sealants used to
assist in the repair of complex, high-risk cardiac and
aortic procedures are not unreasonable to limit bleeding
REVIEW
in certain key situations (eg, left ventricular free-wall
rupture and aortic dissection), but they are associated
with complications that may limit their usefulness in less
high-risk situations.
Several of the hemostatic agents discussed in this
review contain bovine or human-derived components
(Table 2). As with any animal-derived or human plasma-
derived component, immunological reactions may occur
causing severe hypotension, systemic inflammatory re-
sponses, and in rare cases, anaphylaxis. In addition, with
bovine or human-derived plasma components, there is a
small risk of transmission of viral agents, and this is
stated in the manufacturer’s product information. With
any bovine sourced components, there is a risk of trans-
mission of bovine spongiform encephalopathy, but there
is no evidence that such an event has occurred. There has
been a reported case of viral transmission with the use of
an adhesive agent prepared from human plasma [41], but
this pre-dates the current stringent collection and testing
processes, and the current techniques used to reduce
viral load.
 1382 REVIEW BARNARD AND MILLNER
TOPICAL HEMOSTATIC AGENTS FOR USE IN CARDIAC SURGERY
Table 2. Composition of Hemostatic Agents
No Human or Human Plasma
Bovine Component Bovine Component Component
Surgicel, Oxycel Avitene, Colgel, Crosseal/Quixilb
Helitenea
Recothrom ¢ Tisseelc ¡
Omnex ¢ Beriblastc ¡
Coseal ¢ Floseald ¡
e agents while still on cardiopulmonary bypass or while
Arista, Hemostase Thrombin
Hem Con, Closure, CoStasisf
Chitoseal, Celox
BioGlueg
a b
Water-insoluble acid salts of bovine collagen. Human clottable
c
protein, tranexamic acid, and thrombin. Human thrombin and bo-
vine aprotinin. d
Bovine-derived gelatin-based matrix and human-
e
derived thrombin.
f
Thrombin is primarily derived from a bovine
source. Composite of bovine microfibrillar collagen and bovine
thrombin in a calcium chloride buffer mixed with autologous plasma at
the time of surgery. g
Bovine serum albumin and gluteraldehyde
tissue adhesive.
Importantly, surgeons should note that hemostatic
agents carry warning labels detailing contraindications to
their use. One of the important contraindications is the
use of these agents in patients with antibodies against
bovine antibodies. As the study by Weaver and col-
leagues [42] details, 27% of patients exposed to agents
containing bovine thrombin develop these antibodies
and hence should not be exposed to bovine thrombin-
containing agents if future surgery is necessary. The
Society of Thoracic Surgeons Blood Conservation Guide-
line Task Force have recommended that topical hemo-
static agents that use bovine thrombin are not helpful for
blood conservation during cardiopulmonary bypass and
may be potentially harmful [43].
From a practical perspective, the cirmcumstances sur-
rounding the desire to use a hemostatic agent are crucial
to the choice of agent. Hemostatic agents that accelerate
clot formation have the disadvantage that they are only
REVIEW
effective against surgical bleeding when hemorrhage is
minimal, and must therefore be applied before removal
of the cross clamp or while at very low flow on cardio-
pulmonary bypass. In essence this relegates these agents
to either a prophylactic role in which the surgeon must
always anticipate excess hemorrhage in advance, or their
use necessitates reducing flow rates while on cardiopul-
monary bypass or while re-applying the cross clamp, or
with both of the latter two situations. The alternative
option favored by some surgeons in this scenario may be
to use the bovine serum albumin and gluteraldehyde
tissue adhesive with its ability to rapidly form a mechan-
ical seal that is independent of native clotting factors and
which forms even in the presence of some bleeding. The
disadvantages of using the albumin and gluteraldehyde
combination include the reported risks of embolization,
local tissue destruction, and the bovine source of the
albumin with the related risks or anaphylaxis. A further
strategy is to deploy one of the compression hemostatic
agents, such as oxidized regenetrated cellulose gauze.
This offers the advantages of reducing active bleeding,
Ann Thorac Surg
2009;88:1377– 83
providing compression, and providing a scaffold for clot
formation. The disadvantages of using the compression
hemostatic agents is that they may not offer a definitive
solution to prevent further bleeding, they can not en-
hance the process of clot formation in coagulopathic
patients, and they can be responsible for postoperative
infections.
There is a need for caution when choosing hemostatic
using cell salvage systems. Microfibrillar collagen agents
can pass through filtration systems causing end-organ
damage, sponge fabric materials, such as Surgicel, Gel-
foam Sponge, and Hemopad can activate the clotting
cascade causing the cell salvage system to clot off and
liquid agents, such as thrombin and thrombogen may
create a fibrin clot by direct action on fibrinogen, again
clotting off the cell salvage system.
The ideal topical hemostatic agent would be one that
was deployable even against brisk hemorrhage, which
was independent of native clotting mechanisms and
would not pass through salvage filtration systems, and
which was not sourced from a bovine or human origin.
With the development of recombinant thrombin and an
increasing number of agents that do not contain any
animal or human-sourced components, it may be that
there is a shift away from older hemostatic agents.
Preliminary results from animal models show great
promise for Chitosan-containing agents that are inexpen-
sive, bio-absorbable, do not contain any human or bo-
vine-derived components, and have no risk of viral agent
transmission. Clinical trials are required to delineate the
potential benefits of these newer agents and establish
their efficacy in comparison with currently used hemo-
static adjuncts.
References
1. LeNarz LA. Coronary artery bypass graft in abciximab-
treated patients. Ann Thorac Surg 2000;70:S38 – 42.
2. Moulton MJ, Creswell LL, Mackey ME, Cox JL, Rosenbloom
M. Reexploration for bleeding is a risk factor for adverse
outcomes after cardiac operations. J Thorac Cardiovasc
Surg 1996;111:1037– 46.
3. Hall TS, Sines JC, Spotnitz AJ. Hemorrhage related reex-
ploration following open heart surgery: the impact of pre-
operative and post-operative coagulation testing. Cardio-
vasc Surg 2002;10:146 –53.
4. Kozen BG, Kircher SJ, Henao J, Godinez FS, Johnson AS. An
alternative hemostatic dressing: comparison of CELOX,
HemCon, and QuikClot. Acad Emerg Med 2008;15:74 – 81.
5. Scher KS, Coil JA, Jr. Effects of oxidized cellulose and
microfibrillar collagen on infection. Surgery 1982;91:301– 4.
6. Spangler D, Rothenburger S, Nguyen K, Jampani H, Weiss
S, Bhende S. In vitro antimicrobial activity of oxidized
regenerated cellulose against antibiotic-resistant microor-
ganisms. Surg Infect (Larchmt) 2003;4:255– 62.
7. Sirlak M, Eryilmaz S, Yazicioglu L, et al. Comparative study
of microfibrillar collagen hemostat (Colgel) and oxidized
cellulose (Surgicel) in high transfusion-risk cardiac surgery.
J Thorac Cardiovasc Surg 2003;126:666 –70.
8. Robicsek F. Microfibrillar collagen hemostat in cardiac
surgery. J Thorac Cardiovasc Surg 2004;127:1228.
Ann Thorac Surg
2009;88:1377– 83 TOPICAL HEMOSTATIC AGENTS FOR USE IN CARDIAC SURGERY
*9. CoStasis Multi-center Collaborative Writing Committee.
A novel collagen-based composite offers effective hemosta-
sis for multiple surgical indications: Results of a random-
ized controlled trial. Surgery 2001;129:445–50.
10. Schoenecker JG, Hauck RK, Mercer MC, Parker W, Lawson
JH. Exposure to topical bovine thrombin during surgery
elicits a response against the xenogeneic carbohydrate
galactose alpha1-3galactose. J Clin Immunol 2000;20:
434 – 44.
11. Chapman WC, Singla N, Genyk Y, et al. A phase 3,
randomized, double-blind comparative study of the efficacy
and safety of topical recombinant human thrombin and
bovine thrombin in surgical hemostasis. J Am Coll Surg
2007;205:256 – 65.
*12. Lumsden AB, Heyman ER. Prospective randomized study
evaluating an absorbable cyanoacrylate for use in vascular
reconstructions. J Vasc Surg 2006;44:1002–9.
13. Brunkwall J, Ruemenapf G, Florek HJ, Lang W, Schmitz-
Rixen T. A single arm, prospective study of an absorbable
cyanoacrylate surgical sealant for use in vascular recon-
structions as an adjunct to conventional techniques to
achieve haemostasis. J Cardiovasc Surg (Torino) 2007;48:
471– 6.
14. Matras H, Dinges HP, Lassmann H, Mamoli B. [Suture-free
interfascicular nerve transplantation in animal experi-
ments]. Wien Med Wochenschr 1972;122:517–23.
15. Tredree RBW, Debrix I, Eisert A, et al. Evaluating the
difference between fibrin sealants: recommendations from
an international advisory panel of hospital pharmacists.
EJHP Science 2006;12:3–9.
16. Kheirabadi BS, Field-Ridley A, Pearson R, MacPhee M,
Drohan W, Tuthill D. Comparative study of the efficacy of
the common topical hemostatic agents with fibrin sealant in
a rabbit aortic anastomosis model. J Surg Res 2002;106:99 –
107.
17. Busuttil RW. A comparison of antifibrinolytic agents used
in hemostatic fibrin sealants. J Am Coll Surg 2003;197:
1021– 8.
18. Fergusson DA, Hebert PC, Mazer CD, et al. A comparison
of aprotinin and lysine analogues in high-risk cardiac
surgery. N Engl J Med 2008;358:2319 –31.
19. Kjaergard HK, Trumbull HR. Vivostat system autologous
fibrin sealant: preliminary study in elective coronary by-
pass grafting. Ann Thorac Surg 1998;66:482– 6.
20. Velada JL, Hollingsbee DA. Physical characteristics of Vi-
vostat patient-derived sealant. Implications for clinical use.
Eur Surg Res 2001;33:399 – 404.
21. Kjaergard H, Fairbrother J. Controlled clinical studies of
fibrin sealant in cardiothoracic surgery—a review. Eur
J Cardiothorac Surg 1996;10:727–33.
*22. Codispoti M, Mankad PS. Significant merits of a fibrin
sealant in the presence of coagulopathy following paediat-
ric cardiac surgery: randomised controlled trial. Eur J Car-
diothorac Surg 2002;22:200 –5.
23. Lamm P, Adelhard K, Juchem G, et al. Fibrin glue in
coronary artery bypass grafting operations: casting out the
Devil with Beelzebub? Eur J Cardiothorac Surg 2007;32:567–
72.
*24. Hagberg RC, Safi HJ, Sabik J, Conte J, Block JE. Improved
intraoperative management of anastomotic bleeding dur-
ing aortic reconstruction: results of a randomized controlled
trial. Am Surg 2004;70:307–11.
25. Glickman M, Gheissari A, Money S, Martin J, Ballard JL. A
polymeric sealant inhibits anastomotic suture hole bleed-
ing more rapidly than gelfoam/thrombin: results of a ran-
domized controlled trial. Arch Surg 2002;137:326 –32.
*Studies marked with an asterisk are those which have declared commer-
cial sponsorship.
REVIEW BARNARD AND MILLNER 1383
26. Patila T, Jokinen JJ, Salminen J, Kankuri E, Harjula A.
Polyglycolic acid glue does not prevent intrapericardial
adhesions in a short-term follow-up. J Surg Res 2008;
148:181– 4.
*27. Coselli JS, Bavaria JE, Fehrenbacher J, Stowe CL, Macheers
SK, Gundry SR. Prospective randomized study of a protein-
based tissue adhesive used as a hemostatic and structural
adjunct in cardiac and vascular anastomotic repair proce-
dures. J Am Coll Surg 2003;197:243–53.
28. Lemaire SA, Ochoa LN, Conklin LD, et al. Nerve and
conduction tissue injury caused by contact with BioGlue.
J Surg Res 2007;143:286 –93.
29. LeMaire SA, Carter SA, Won T, Wang X, Conklin LD,
Coselli JS. The threat of adhesive embolization: BioGlue
leaks through needle holes in aortic tissue and prosthetic
grafts. Ann Thorac Surg 2005;80:106 –11.
30. LeMaire SA, Schmittling ZC, Coselli JS, et al. BioGlue
surgical adhesive impairs aortic growth and causes anasto-
motic strictures. Ann Thorac Surg 2002;73:1500 – 6.
31. Devbhandari MP, Chaudhery Q, Duncan AJ. Acute intra-
operative malfunction of aortic valve due to surgical glue.
Ann Thorac Surg 2006;81:1499 –500.
32. Karimi M, Kerber RE, Everett JE. Mechanical aortic valve
malfunction: an intraoperative BioGlue complication.
J Thorac Cardiovasc Surg 2005;129:1442–3.
33. Kazui T, Washiyama N, Bashar AH, et al. Role of biologic
glue repair of proximal aortic dissection in the development
of early and midterm redissection of the aortic root. Ann
Thorac Surg 2001;72:509 –14.
34. Ngaage DL, Edwards WD, Bell MR, Sundt TM. A caution-
ary note regarding long-term sequelae of biologic glue.
J Thorac Cardiovasc Surg 2005;129:937– 8.
*35. Oz MC, Cosgrove DM, 3rd, Badduke BR, et al. Controlled
clinical trial of a novel hemostatic agent in cardiac surgery.
The Fusion Matrix Study Group. Ann Thorac Surg 2000;69:
1376 – 82.
36. Malette WG, Quigley HJ, Gaines RD, Johnson ND, Rainer
WG. Chitosan: a new hemostatic. Ann Thorac Surg 1983;36:
55– 8.
37. Okamoto Y, Yano R, Miyatake K, Tomohiro I, Shigemasa Y,
Minami S. Effects of chitin and chitosan on blood coagula-
tion. Carbohydr Polym 2003;53:337– 42.
38. Rao SB, Sharma CP. Use of chitosan as a biomaterial:
studies on its safety and hemostatic potential. J Biomed
Mater Res 1997;34:21– 8.
39. Millner RW, Lockhart AS, Bird H, Alexiou C. A new
hemostatic agent: initial life-saving experience with Celox
REVIEW
(chitosan) in cardiothoracic surgery. Ann Thorac Surg 2009;
87:e13– 4.
40. Weaver FA, Hood DB, Zatina M, Messina L, Badduke B.
Gelatin-thrombin-based hemostatic sealant for intraopera-
tive bleeding in vascular surgery. Ann Vasc Surg 2002;16:
286 –93.
41. Wilson SM PP, Donegan EA. HIV1 transmission following
the use of cryoprecipitated fibrinogen as gel/adhesive.
Transfusion 1991;31:51.
*42. Weaver FA, Lew W, Granke K, Yonehiro L, Delange B,
Alexander WA. A comparison of recombinant thrombin to
bovine thrombin as a hemostatic ancillary in patients un-
dergoing peripheral arterial bypass and arteriovenous graft
procedures. J Vasc Surg 2008;47:1266 –73.
43. Ferraris VA, Ferraris SP, Saha SP, et al. Perioperative blood
transfusion and blood conservation in cardiac surgery: the
Society of Thoracic Surgeons and The Society of Cardiovas-
cular Anesthesiologists clinical practice guideline. Ann
Thorac Surg 2007;83:S27– 86.
*44. Rousou J, Levitsky S, Gonzalez-Lavin L, et al. Randomized
clinical trial of fibrin sealant in patients undergoing rester-
notomy or reoperation after cardiac operations. A multi-
center study. J Thorac Cardiovasc Surg 1989;97:194 –203.