themenschwerpunkt
Wien Med Wochenschr (2011) 161/3–4: 68–72
DOI 10.1007/s10354-011-0878-6

Springer-Verlag 2011
Printed in Austria
Thrombosis prophylaxis in critically ill patients
Dietmar Fries
Department of General and Surgical Intensive Care Medicine, Medical University of Innsbruck,
Innsbruck, Austria
Received November 10, 2010, accepted December 14, 2010
Thromboseprophylaxe bei kritisch Kranken
Zusammenfassung. Die Häufigkeit von tiefen Beinvenen-
thrombosen bei Intensivpatienten wird in Abhängigkeit von der
Grunderkrankung mit bis zu etwa 60 % angegeben. Die aktuellen
Empfehlungen sind, obwohl die Datenlage in diesem selektio-
niertem Krankengut sehr spärlich ist, eindeutig: Die Surviving
Sepsis Campaign Richtlinie fordert eine prophylaktische Anti-
koagulation bei Patienten mit schwerer Sepsis bzw. mit sep-
tischen Schock solange keine Kontraindikationen vorliegen.
Dieser Artikel beschäftigt sich mit Risikofaktoren für das Auf-
treten einer Thromboembolie beim kritisch Kranken und mit
Möglichkeiten der physikalischen und medikamentösen Antikoa-
gulation. Periphere Vasokonstriktion, Schock sowie die Verab-
reichung von Katecholaminen, Ödem mit teilweise massiven
Flüssigkeitsansammlungen als Folge der Flüssigkeitsretention
v. a. im Bindegewebe sowie eine mitunter stark eingeschränkte
Organfunktion lassen die Wirkung von subkutan verabreichten
niedermolekularen Heparinen nicht sicher vorhersagen. Dieser
Beitrag beschäftigt sich auch mit dem Problem der Heparin-
Resistenz und den pathophysiologischen Ursachen. Die konti-
nuierliche intravenöse Applikation neuer Antikoagulantien
könnte hier eine sinnvolle Alternative sein. Für den direkten
Thrombininhibitor Argatroban gilt, dass je ausgeprägter die Er-
krankung bzw. je höher der SAPS-II-Score, desto niedriger ist die
erforderliche Dosis.
Schlüsselwörter: Venöse Thrombose, Thromboseprophylaxe,
Intensivbehandlung
Summary. Incidence of deep vein thrombosis in critically ill
patients depends on the underlying disease but may be as high as
60%. The Surviving Sepsis Campaign clearly recommends ad-
ministering anticoagulation in the absence of specific contra-
indications in patients with severe sepsis or septic shock.
The article discusses risk factor for thromboembolic events in
critical illness as well as means of non-pharmacologic and phar-
macologic thrombosis prophylaxis. Peripheral vasoconstriction,
edema, shock, and administration of catecholamines may reduce
Correspondence: Dietmar Fries, Ph.D., Department of Anaesthesiol-
ogy and Critical Care Medicine, Medical University of Innsbruck,
Anichstraße 35, 6020 Innsbruck, Austria.
Fax: þþ43-512-504-22490, E-mail: dietmar.fries@i-med.ac.at
68
the bioavailability and efficacy of subcutaneous administration of
low molecular weight heparin. This article further elaborates on
the problem and pathophysiology of heparin resistance. Contin-
uous intravenous administration of new anticoagulants may be a
promising alternative to indirect anticoagulants. Severity of ill-
ness and SAPS II-score determine dosing of the direct thrombin
inhibitor argatroban which needs to be about 10-times lower than
in patients without critical illness.
Key words: Venous thrombosis, drug therapy, critical illness
Introduction
Despite the great advances made in anticoagula-
tion therapy in recent years, thrombosis and pulmonary
embolism remain preventable complications, especial-
ly in critically ill patients, and significantly influence
morbidity and mortality. Only few epidemiologic data
on the occurrence and prevention of thromboembolic
complications are available for this selected patient
population. The occurrence of asymptomatic deep vein
thrombosis in severely injured patients is estimated at
approx. 60% [1].
Internal medicine and general surgical intensive
care patients without prophylactic anticoagulation, on
the other hand, appear to develop a thrombosis in only
about 30% of cases, with the reported statistics differing
greatly [2]. Other authors report the incidence of deep
vein thrombosis in such intensive care patients at
13–31% [3]. Despite prophylactic anticoagulation drug
therapy, the occurrence of thromboses in these patients
is of great significance. Despite prophylactic anticoa-
gulation, deep vein thrombosis can be expected in
about 10% of patients staying at the ICU for more than
three days [4].
Trauma intensive care patients presumably have
the greatest risk of suffering a thrombosis and/or em-
bolism. In 58% of trauma patients who received no

Springer-Verlag 3–4/2011 wmw
 themenschwerpunkt

Tab. 1: Thromboembolic complications and DVT in critical ill patients with and without antithrombotic
prophylactic treatment
No Antithrombotics (%)
antithrombotics (%)

Moser et al. [11] No treatment 9 –

Cade [12] Placebo/UFH 29 13

Hirsch et al. [2] Individual decision 32 40

Marik et al. [13] Individual decision 25 7

Kapoor et al. [14] Placebo/UFH 29 13

Fraisse et al. [15] Placebo/LMWH 28 15

DVT deep vein thrombosis, UFH unfractionated heparin, LMWH low molecular weight heparin.

thrombosis prophylaxis, a thrombosis was diagnosed
between the second and third week [5]. However, the
course of these patients has been asymptomatic largely
[5]. Two doses of 5000 IU UFH administered subcuta-
neously can reduce the incidence of thrombosis in such
patients by about 20%. LMWH can cause an additional
30% reduction in incidence [6]. Fears that a significant
bleeding episode could occur on prophylactic antic-
oagulation are generally not justified, provided initial
hemostasis is achieved. In patients on UFH a relevant
bleeding was detected in some 0.5% of cases, and
surgical intervention was necessary in only a few cases.
However, the incidence appears to be somewhat in-
creased in patients on LMWH [6].
Risk factors that favor the occurrence of a throm-
boembolic complication in critically ill patients are
[2, 7–9]:
1. age,
2. deep vein thrombosis in the patient’s anamnesis,
3. severe trauma,
4. long ICU stay,
5. mechanical ventilation,
6. analgosedation/relaxation,
7. emergency medical interventions,
8. placement of a central venous catheter in the
v. femoralis,
9. no thrombosis prophylaxis.
For the sake of simplicity it can be said that the risk of
suffering a thromboembolic complication at the ICU is
directly related to the severity of the underlying disor-
der and the degree of immobilization.
The recommendations and guidelines currently
in place are clear despite the fact that only very few data
are available for this selected patient population. The
Surviving Sepsis Campaign guideline demands prophy-
lactic anticoagulation in patients with severe sepsis or
suffering from septic shock, as long as there are no
contraindications [10]. Whether low molecular weight
or unfractionated heparin is to be administered and
whether it is given subcutaneously or intravenously was
not mentioned there.
Prophylactic anticoagulation drug therapy gen-
erally appears to be very effective in critically ill pa-
tients. Administration of LMWH or UFH can reduce
the incidence of deep vein thrombosis by approx. 50%
(Tab. 1).
Unfractionated versus low molecular
weight heparin in intensive care
medicine
Both of these preparations are indirect thrombin
inhibitors, meaning the patient’s antithrombin level
must be high enough in order for them to be effective.
Only a few studies have compared the efficacy of UFH
with that of LMWH. Seen from a practical clinical
standpoint, the advantages for the administration of
UFH are obvious:
1. effect of UFH can be completely reversed using
protamine sulfate,
2. simple and easily available monitoring of UFH
using apt,
3. low-cost therapy,
4. UFH is backed up by long years of experience
with it.
There are, however, several disadvantages:
1. thrombocyte activation,
2. higher incidence of heparin-induced thrombocyto-
penia type 2 (HIT 2),
3. apTT monitoring can be distorted by various con-
founding factors.

wmw 3–4/2011
Springer-Verlag Fries – Thrombosis prophylaxis in critically ill patients 69

 themenschwerpunkt
A very few comparative investigations demonstrated an
advantage for LMWH:
Geerts et al. compared the effect of 2  30 mg
s.c. enoxaparin sodium with the administration of
2  5000 IU UFH s.c. in trauma patients. In that study
enoxaparin sodium reduced the incidence of thrombo-
sis by 30%, and a statistically not significant trend was
detected in enoxaparin-treated patients with the regard
to bleeding complications [6]. Patients with spinal cord
injuries and motor paralysis also showed an advantage
for the administration of LMWH [16]. Also in neurosur-
gic patients the additional administration of LMWH
reduced the incidence of thrombosis by 26% as com-
pared to compression hosiery [17].
The decision whether an ICU patient should be
treated with UFH or LMWH can, however, not be
answered from large randomized studies, but must
be decided on a case-to-case basis depending on the
clinical situation. If the thrombosis risk is accompanied
by a bleeding risk or if an additional surgical procedure
is planned for the near future, the attending intensive
care physician will decide for UFH therapy, because it
can be reversed, if needed. Intensive care patients with
severely impaired renal function are also more likely to
receive UFH in order to avoid a possible overdose of
LMWH and possible bleeding complications. However,
if the bleeding risk in an intensive care patient is
negligible, s.c. LMWH therapy can be commenced. At
our department this is accompanied by daily anti-Xa
monitoring, primarily to detect any possible overdos-
age as early as possible.
However, ICU physicians have to take into ac-
count that underdosage of LMWH with inadequate low
ant-Xa levels may result in thromboembolic complica-
tions [18]. To achieve an adequate anti-Xa response, a
dosage of about 60 mg/d enoxaprin was found to be
effective in critically ill patients in a double blinded
randomized trial [19]. On the other hand, administra-
tion of high amounts of LMWH can result in accumu-
lation with the appearance of bleeding, especially in
patients with acute or chronic renal failure.
Intravenous versus subcutaneous
administration
Subcutaneous administration of medication in
intensive care patients must always be given critical
consideration. Peripheral vasoconstriction, shock and
catecholamine administration, edema, sometimes
massive as the result of fluid retention above all in
connective tissue, as well as sometimes strongly im-
70 Fries – Thrombosis prophylaxis in critically ill patients
paired organ function do not permit the effect of s.c.
administration of preparations to be predicted with
certainty.
A prospective study demonstrated that subcuta-
neous administration of 40 mg enoxaparin sodium
once daily in critically ill ICU patients was able to
induce the recommended anti-Xa values in only a few
cases [20]. Above all, patients with high bodyweight and
pronounced multiorgan failure responded only insuffi-
ciently to subcutaneous administration of exoxaparin
sodium [21]. A different study conducted in intensive
care patients looked at the subcutaneous administra-
tion of certoparin once or twice daily. In that study,
however, low antithrombin levels were associated with
poor response to certoparin [21].
Heparin resistance – an underestimated
problem in intensive care medicine?
The clinical effectiveness of heparin is dependent
on achieving a defined anticoagulant effect. In vivo,
there is a huge variation in response to a fixed dose of
heparin between individuals. In this context, a phe-
nomenon known as “heparin resistance” should be
taken into consideration, in particular postoperatively
and in the setting of cardiac bypass surgery. Among risk
factors for developing heparin resistance, decreased
antithrombin (AT) levels represent one of the greatest
risks. Further on, binding of heparin to plasma proteins
including platelet factor 4 (PF4), fibrinogen, factor VIII,
and histidine-rich glycoprotein may cause heparin re-
sistance. As many heparin-binding proteins are acute-
phase reactants, the phenomenon of heparin resistance
is often encountered in acutely ill patients, in patients
with malignancy, and during peri- or post-partum per-
iods. In addition to altered mechanisms of heparin
clearance, heparin resistance has also been associated
with drug-induced causes including aprotinin and ni-
troglycerin, although the latter is controversial. Its as-
sociation with low antithrombin levels is also a point of
heated discussion in the literature. Heparin therapy
produces a decrease in circulating antithrombin that
is independent on the initial dose, is detectable after
1 day, and peaks after 2–4 days. Cessation of therapy
leads to a normalization of AT levels, but it is not clear if
this reduction of AT contributes to the heparin resis-
tance seen in patients undergoing treatment for venous
thromboembolism, or in patients undergoing cardiac
bypass surgery.
For prophylactic administration of heparin, clear
cut-off values are not defined within the literature to

Springer-Verlag 3–4/2011 wmw
 themenschwerpunkt

determine the presence of heparin resistance. In clini- reliable anticoagulant effect independent of the level
cal practice, dosages of more than 28,000 U/24 h of acute phase reactants.
(1.200 U per hour) heparin or above seem to be indica-
tive. For the treatment of thromboembolic complica-
tions, heparin resistance is defined as the need for Direct thrombin inhibitors in intensive
more than 35,000 U/24 h (1.500 U per hour) to prolong care medicine
the activated partial thromboplastin time (apTT) into
the therapeutic range. During cardiac bypass proce- Lepirudin and argatroban count among the direct
dures, the definition of heparin resistance is based on thrombin inhibitors. In contrast to the indirect throm-
the activated clotting time (ACT), with at least one ACT bin inhibitors (UFH, LMWH), their effect is indepen-
less than 400 s after heparinization and/or the need for dent of antithrombin. Argatroban is being increasingly
exogenous antithrombin administration. An inade- used for other than the approved indications (throm-
quate response to heparin, known as heparin resis- bosis prophylaxis for HIT 2). Argatroban is above all
tance, has been reported in up to 22% of patients used as anticoagulation in continuous venovenous re-
undergoing CPB. nal replacement therapy, but also for patients with a
In summary, in the treatment and prophylaxis of high risk for thromboembolism and poor response to
venous thromboembolism the phenomenon of heparin heparin [22]. However, it must be remembered that in
resistance is responsible for insufficient antithrombotic intensive care patients argatroban dosage must be
prophylaxis or therapy. In this context, the use of direct calculated on a patient-to-patient basis. Contrary to the
thrombin inhibitors may be useful. Since thrombin initial manufacturer’s recommendation, an up to 10-
inhibitors bind directly to the active site of thrombin, fold smaller dose can suffice to produce a satisfactory
their pharmacologic action does not require endoge- anticoagulatory effect [23]. In critically ill patients with
nous AT. Because of its small molecular size, argatro- heparin-induced thrombocytopenia type II and neces-
ban is also effective in inhibiting clot-bound thrombin, sity for hemofiltration, critical illness scores have been
which is resistant to neutralization by the heparin/AT able to predict the required argatroban maintenance
complex. Furthermore, argatroban exhibits a relatively dose for anticoagulation [24]. The more pronounced
low protein binding in plasma which results in a more the disorder or the higher the Simplified Acute

r = –0.81 r = –0.80
Argatroban (µg/kg x min)

Argatroban (µg/kg x min)

1.5 p < 0.001 1.5 p < 0.001

n = 30 n = 30

1.0 1.0

0.5 0.5

0.01 0.01

0 0
0 15 20 25 30 35 0 30 40 50 60
APACHE-II Score (Punkte) SAPS-II Score (Punkte)

r = 0.89 r = –0.90
p < 0.001 p < 0.001
Argatroban (µg/kg x min)

Argatroban (µg/kg x min)

1.5 n = 30
n = 23
1.5

1.0
1.0

0.5
0.5
0.01

0 0 24
0 6 8 10 12 14 16 18 20 22 0 9 12 15 18 21
ICG-PDR (%/min) SOFA Score (Punkte)
Fig. 1: Acute Physiology and Chronic Health Evaluation (APACHE-II), Simplified Acute Physiology Score (SAPS-II), Sequential Organ Failure Assessment Score
(SOFA) or indocyanine green plasma disappearance rate (ICGPDR) predict the required argatroban maintenance dosage for anticoagulation. These predictors
identify decreased argatroban dosing [24]

wmw 3–4/2011
Springer-Verlag Fries – Thrombosis prophylaxis in critically ill patients 71

 themenschwerpunkt
Physiology Score II (SAPS-II) score or the Acute Physi-
ology and Chronic Health Evaluation II (APACHE-II)
score, the more caution is warranted when estimating
the dosage [24]:
argatroban dose½mg=kg  min ¼ 2:06  0:03  SAPS II
argatroban dose½mg=kgmin
¼ 2:15  0:06  APACHE II
To transfer this observation presented in Fig. 1 into
clinical practice, there is a rule of thumb: The more
sick the patient is, the less agratroban has to be ad-
ministered in order to avoid accumulation with over-
dosing and bleeding complications. Bleeding
complications following an overdosing of argatroban
may be difficult to manage e.g. because there is no
antidote available.
Physical thrombosis prophylaxis
The efficacy of compression hosiery and of
pneumatic compression systems has been demon-
strated above all in neurosurgical intensive care
patients. The best results, however, were achieved
with additional administration of enoxaparin sodi-
um, with bleeding risk also increasing when LMWH
was given [17].
Conclusion
Prophylactic anticoagulation in critically ill pa-
tients at the ICU differs in many aspects from that in
other patients. Dosage must be calculated on a patient-
to-patient basis according to the clinical circumstances.
Laboratory parameters are often difficult to interpret
and can mask the anticoagulatory effect of, for example,
heparin. New and very potent antithrombotic agents
are being increasingly used at the ICU, but the evi-
dence-based data available in connection with them
are still very meager. Physical thrombosis prophylaxis
should be considered in critically ill patients at high risk
for bleeding and thrombosis.
Conflict of interest
The author received lecture fee from Mitsubishi
Pharma.
72 Fries – Thrombosis prophylaxis in critically ill patients
References
[1] Kudsk KA, Fabian TC, Baum S, Gold RE, Mangiante E, Voeller G.
Silent deep vein thrombosis in immobilized multiple trauma pa-
tients. Am J Surg, 158(6): 515–519, 1989.
[2] Hirsch DR, Ingenito EP, Goldhaber SZ. Prevalence of deep venous
thrombosis among patients in medical intensive care. JAMA, 274(4):
335–337, 1995.
[3] Geerts W, Cook D, Selby R, Etchells E. Venous thromboembolism
and its prevention in critical care. J Crit Care, 17(2): 95–104, 2002.
[4] Cook D, Crowther M, Meade M, et al. Deep venous thrombosis in
medical-surgical critically ill patients: prevalence, incidence, and risk
factors. Crit Care Med, 33(7): 1565–1571, 2005.
[5] Geerts WH, Code KI, Jay RM, Chen E, Szalai JP. A prospective study of
venous thromboembolism after major trauma. N Engl J Med, 331(24):
1601–1606, 1994.
[6] Geerts WH, Jay RM, Code KI, et al. A comparison of low-dose heparin
with low-molecular-weight heparin as prophylaxis against venous
thromboembolism after major trauma. N Engl J Med, 335(10):
701–707, 1996.
[7] Harris LM, Curl GR, Booth FV, Hassett JM Jr, Leney G, Ricotta JJ.
Screening for asymptomatic deep vein thrombosis in surgical inten-
sive care patients. J Vasc Surg, 26(5): 764–769, 1997.
[8] Joynt GM, Kew J, Gomersall CD, Leung VY, Liu EK. Deep venous
thrombosis caused by femoral venous catheters in critically ill adult
patients. Chest, 117(1): 178–183, 2000.
[9] Marik PE, Andrews L, Maini B. The incidence of deep venous
thrombosis in ICU patients. Chest, 111(3): 661–664, 1997.
[10] Dellinger RP, Levy MM, Carlet JM, et al. Surviving Sepsis Campaign:
international guidelines for management of severe sepsis and septic
shock: 2008. Crit Care Med, 36(1): 296–327, 2008.
[11] Moser KM, LeMoine JR, Nachtwey FJ, Spragg RG. Deep venous
thrombosis and pulmonary embolism. Frequency in a respiratory
intensive care unit. JAMA, 246(13): 1422–1424, 1981.
[12] Cade JF. High risk of the critically ill for venous thromboembolism.
Crit Care Med, 10(7): 448–450, 1982.
[13] Marik PE, Andrews L, Maini B. The incidence of deep venous
thrombosis in ICU patients. Chest, 111(3): 661–664, 1997.
[14] Kapoor A, Tewari S. Re Sandoval, et al. Thromboembolic stroke
complicating balloon angioplasty of a vein graft. Catheter Cardiovasc
Interv, 46(1): 119–121, 1999.
[15] Fraisse F, Holzapfel L, Couland JM, et al. Nadroparin in the preven-
tion of deep vein thrombosis in acute decompensated COPD. The
Association of Non-University Affiliated Intensive Care Specialist
Physicians of France. Am J Respir Crit Care Med, 161(4 Pt 1):
1109–1114, 2000.
[16] Green D, Lee MY, Lim AC, et al. Prevention of thromboembolism
after spinal cord injury using low-molecular-weight heparin.
Ann Intern Med, 113(8): 571–574, 1990.
[17] Agnelli G, Piovella F, Buoncristiani P, et al. Enoxaparin plus com-
pression stockings compared with compression stockings alone in
the prevention of venous thromboembolism after elective neurosur-
gery. N Engl J Med, 339(2): 80–85, 1998.
[18] Robinson S, Zincuk A, Strøm T, et al. Enoxaparin, effective dosage for
intensive care patients: double-blinded, randomised clinical trial.
Crit Care, 14(2): R41, 2010.
[19] Malinoski D, Jafari F, Ewing T, Ardary C, et al. Standard prophylactic
enoxaparin dosing leads to inadequate anti-Xa levels and increased
deep venous thrombosis rates in critically ill trauma and surgical
patients. J Trauma, 68(4): 874–880, 2010.
[20] Mayr AJ, Dünser M, Jochberger S, et al. Antifactor Xa activity in
intensive care patients receiving thromboembolic prophylaxis with
standard doses of enoxaparin. Thromb Res, 105(3): 201–204, 2002.
[21] Jochberger S, Mayr V, Luckner G, et al. Antifactor Xa activity in
critically ill patients receiving antithrombotic prophylaxis with stan-
dard dosages of certoparin: a prospective, clinical study. Crit Care,
9(5): R541–R548, 2005.
[22] Link A, Girndt M, Selejan S, Mathes A, Böhm M, Rensing H. Arga-
troban for anticoagulation in continuous renal replacement therapy.
Crit Care Med, 37(1): 105–110, 2009.
[23] Beiderlinden M, Treschan TA, Görlinger K, Peters J. Argatroban
anticoagulation in critically ill patients. Ann Pharmacother, 41(5):
749–754, 2007.
[24] Link A, Girndt M, Selejan S, et al. Argatroban for anticoagulation in
continuous renal replacement therapy. Crit Care Med, 37(1):
342–343, 2009.

Springer-Verlag 3–4/2011 wmw