Early Postoperative Anticoagulation After
Mechanical Valve Replacement: A Systematic
Review
Alexander Kulik, MD, Fraser D. Rubens, MD, MS, Philip S. Wells, MD, MS,
Clive Kearon, MB, PhD, Thierry G. Mesana, MD, PhD, Judith van Berkom, BA, and
B-Khanh Lam, MD, MPH
Division of Cardiac Surgery, University of Ottawa Heart Institute, Division of Clinical Hematology, Ottawa Hospital, Ottawa, and
the Department of Medicine, McMaster University, Hamilton, Ontario, Canada
The optimal approach to early postoperative anticoagu-
lation after mechanical valve implantation remains con-
troversial. This review article examines the pathogenesis
of thrombus formation and the different strategies for
early postoperative anticoagulation. The most commonly
reported anticoagulation regimens had the after esti-
mates of early postoperative thromboembolism and hem-
orrhage: oral anticoagulation alone (0.9%, 3.3%); oral
anticoagulation with intravenous unfractionated heparin
(1.1%, 7.2%); and oral anticoagulation with low molecular
weight heparin (0.6%, 4.8%). Although intravenous hep-
arin may be associated with a higher incidence of hem-
orrhage, a randomized trial is needed to provide the best
evidence regarding early postoperative anticoagulation
after mechanical valve implantation. Nearly four decades
REVIEWS
have passed since the first mechanical prosthetic valves
were implanted. Frequent thromboembolic complica-
tions with the first mechanical valves led to recommen-
dations of universal anticoagulation for these patients.
Since then, several design changes and modifications
have been made to improve the longevity, hemodynam-
ics, and thrombogenicity of newer generation mechanical
valves. With improved blood flow, less stasis, and less
thrombogenic materials, lower rates of thromboembo-
lism have been reported [1]. Despite these advances
however, thromboembolism and anticoagulant-related
Material and Methods
A computerized literature search for abstracts was
performed in accordance with recommendations of
the Cochrane collaboration using MEDLINE, EMBASE,
CINAHL, BIOSIS, EICompendex, SIGLE, and the Co-
chrane Library from the earliest available date to June
2004. The initial keywords and MESH terms were peri-
operative care, perioperative, anticoagulants, heparin,
warfarin, LMWH, low molecular weight heparin, aspirin,
ASA, warfarin, coumadin, heart valve prosthesis, heart
valve prosthesis implantation, heart valve replacement,
Address correspondence to Dr Lam, University of Ottawa Heart Institute,
40 Ruskin Street, Suite H3404, Ottawa, Ontario, Canada K1Y 4W7; e-mail:
bklam@ottawaheart.ca.
2006 by The Society of Thoracic Surgeons
Published by Elsevier Inc
bleeding continue to account for 75% of all complications
after mechanical valve replacement [2]. Occurring most
commonly within six months after implantation [2], these
complications can adversely affect mortality and quality
of life. Furthermore, the threat of their occurrence creates
a psychological burden for each patient with a mechan-
ical valve. The need for life-long anticoagulation in
patients with mechanical valves is not in dispute, and the
perioperative management of anticoagulation during
non-cardiac surgery has been reviewed extensively [3].
However, the approach to early postoperative anticoagu-
lation after mechanical valve implantation is still a mat-
ter of debate. The optimal intensity and timing of anti-
coagulation to prevent early thromboembolism after
valve replacement surgery without postoperative bleed-
ing complications is unknown. Hence, many anticoagu-
lation protocols have been proposed, but a lack of con-
sensus remains. The objectives of this study were (1) to
reexamine the pathogenesis of thrombus formation and
the need for anticoagulation; (2) to critically review the
literature on early postoperative anticoagulation strate-
gies; and (3) provide an estimate of the incidence of
bleeding and thromboembolism for each approach to
early postoperative anticoagulation.
(Ann Thorac Surg 2006;81:770 81)
2006 by The Society of Thoracic Surgeons
mechanical valve, mitral valve replacement, thrombosis,
bleeding, hemorrhage, thromboembolism, thrombus,
bleed, clot, and heart valves [surgery]. In addition, an
Internet search was conducted, and sources for practice
guidelines and gray literature were explored. A total of
244 citations were obtained, and relevant manuscripts
were reviewed. The reference list of each pertinent article
was assessed, and the Science Citation Index was exam-
ined for cited references of selected articles. Original
manuscripts and review articles focusing on early
postoperative anticoagulation after mechanical valve im-
plantation were included for evaluation. No language
restrictions were applied. Studies dealing with the peri-
operative anticoagulation of mechanical valves during
noncardiac surgery were specifically excluded. The term
0003-4975/06/$32.00
doi:10.1016/j.athoracsur.2005.07.023
Ann Thorac Surg
2006;81:770 81
early postoperative was defined as less than 30 days
from valve implantation. As per the Guidelines for Re-
porting Morbidity and Mortality after Cardiac Valvular
Operations, thromboembolism was defined as an em-
bolic event that occurred in the absence of infection, after
the immediate operative period, and a bleeding event
was defined as an episode of major internal or external
bleeding that caused death, permanent injury, or re-
quired transfusion [4].
Using the search strategy described above, it was
evident that very few studies directly compared early
postoperative anticoagulation strategies [59], and a
meta-analysis was not possible because of the lack of
data. In order to provide a qualitative comparison of
anticoagulation strategies, an additional MEDLINE
search was conducted to obtain data regarding the risk of
thromboembolism and bleeding associated with each of
the different approaches to early anticoagulation. This
second search focused on randomized controlled trials,
case series, and case-control studies reporting the results
of the most common prosthetic mechanical valves cur-
rently available. The initial keywords and MESH terms
were St. Jude Medical, Medtronic-Hall, Omnicarbon,
Carbomedics, ATS valve, On-x valve, mechanical valve,
anticoagulation, hemorrhage, bleeding, thromboembo-
lism, and human. Manuscripts reporting the experience
of mechanical valves in the aortic and/or mitral position
in adult patients were included in this review. Series
specifically focused on the outcomes of children or redo
operations were excluded, and language restrictions
were not applied. A total of 141 citations were obtained,
and relevant manuscripts were reviewed by two authors
(AK and BKL). The incidences of early postoperative
bleeding and thromboembolism were extracted from
each manuscript (absolute rates), as well as rates of
perioperative mortality caused by hemorrhage or throm-
boembolism. Studies that did not clearly state the early
postoperative anticoagulation strategy were excluded
from further analysis. These data were summarized ac-
cording to the three most common anticoagulation strat-
egies to yield an approximate estimate of the risk of
bleeding and embolism associated with each approach
(Table 1).
Results
Mechanical Valve Thromboembolism and
Anticoagulation
PATHOGENESIS. The pathogenesis of prosthetic valve
thromboembolism is a complex phenomenon, occurring
through an interaction of a variety of prosthesis-related
and patient-related factors. The pathologic events lead-
ing to thromboembolism begin immediately after sur-
gery. Damaged perivalvular tissue and deposition of
fibrinogen on the valve surface activate platelets as soon
as blood starts flowing across the valve. This leads to
immediate platelet adhesion and aggregation [1], and
within 24 hours after surgery, platelet deposition on the
Dacron sewing ring can be imaged radiographically [10].
REVIEW KULIK ET AL 771
REVIEW OF EARLY POSTOPERATIVE ANTICOAGULATION
However, these platelets are highly sensitive to the shear
forces generated by the mechanical valve and are prone
to continued activation and destruction. With continuing
cycles of platelet aggregation, patients with mechanical
valves have shortened platelet survival, a marker of
thromboembolic risk [11].
Coagulation factors are also directly activated after
valve implantation, leading to further clot formation as a
result of the inherent thrombogenicity of the prosthetic
material (suture material, Dacron sewing ring, struts, and
hinge points) and sites of denuded tissue (valve excision
site) [12]. Transprosthetic turbulent flow leads to regional
increases in shear stress, structurally damaging the en-
docardium, causing a loss of local resistance to thrombo-
sis. Recirculation areas on the outflow side of the pros-
thesis create flow stagnation, trapping damaged platelets
and activated factors [13]. This provides an ideal milieu
for thrombus formation and subsequent embolization.
Thrombin can also be formed on platelet membranes
after their activation, further promoting the organization
and growth of platelet fibrin thrombus [14]. In sum, the
particularly high thromboembolic risk that is present
shortly after valve implantation involves both coagula-
tion and platelet activation due to turbulent flow across
the valve and the thrombogenicity of prosthetic material
and denuded tissue.
EARLY THROMBOEMBOLIC RISK AFTER MECHANICAL VALVE
IMPLANTATION. The insertion of a large artificial device in
contact with the bloodstream exposes the patient to a
continual risk of valve thrombosis and embolism. This
REVIEWS
risk is proportional to the surface area of the foreign
material, which is in contact with blood, making patients
with mitral valve prostheses more prone to thromboem-
bolic complications [11]. In general, a recent mechanical
valve implantation is a strong risk factor for thromboem-
bolic complications [15, 16], especially in the first three to
six months after surgery [17, 18]. The reasons for this are
threefold: first, the pathologic sequelae of the patients
inherent valvular disease (atrial fibrillation, dilated left
atrium, and dilated left ventricle) may predispose to
areas of stasis and thrombus formation.
Second, the increased thromboembolic risk early after
mechanical valve implantation may reflect incomplete
endothelial proliferation on the raw intracardiac surfaces,
sewing ring, and suture knots in the initial postoperative
period [1, 12, 18, 19]. The presence of an endothelial
lining on the valve surface effectively prevents thrombus
formation, but more than one year may be required after
implantation for this in-growth to fully form [20, 21].
However, in the absence of endothelial organization, the
development of a mature platelet fibrin coating on the
valve surface may also be favorable in terms of non-
thrombogenicity and explain the absence of thrombotic
deposits in explanted valves in which tissue in-growth is
incomplete [20, 21]. Therefore, it is believed that the lack
of host endothelial cell in-growth and mature platelet
fibrin coating on the valvular surface in the postoperative
period may promote early thrombus formation and con-
tribute to the elevated early thromboembolic risk.
 REVIEWS

772
Table 1. Bleeding and Thromboembolic Rates With Three Common Anticoagulation Protocols
Absolute Bleeding
Study TE Mortality TE Bleeding Mortality Bleeding

REVIEW OF EARLY POSTOPERATIVE ANTICOAGULATION

REVIEW
Publication Type Therapy Study Description Absolute TE rate rate Description Rate Rate Description

Oral Coumadin with

Subcutaneous Heparin

KULIK ET AL
Ageno W, Am J Cardiol RCT Oral coumadin RCT of 2.5 mg 0/197 (0%) 0/197 (0%)
2001 [5] fixed dose
coumadin
versus 5 mg
load (58%
mechanical
valve patients)
Akins CW, Ann Thorac CS Oral coumadin on day 391 MH 1/391 (0.3%) 1/391 (0.3%) Gl bleed
Surg. 1996 [59] one, with IV
heparin or dextran
on day 5 if INR
subtherapeutic
Anttila V, Scan CC Oral coumadin on 43 MH vs 48 SJM 2/91 (2.2%) 2 CVA 5/91 (5.5%)
Cardiovasc 2002 [60] POD2 or POD3
Bernal JM, Ann Thorac CS Oral coumadin on 1049 CM 0/1049 (0%) 15/1049 (1.4%)
[61] Surg, 1998 POD2
Dalrymple-Hay MJR, CS Oral coumadin POD0; 1350 CM 6/1350 (0.4%) 6 CVA 8/1350 (0.6%)
JHVD 2000 [62] IV heparin on POD3
if INR 2
de la Fuente A, JHVD CC Oral Coumadin on 99 CM vs 93 0/99 (0%) NA 1/99 (1%)
2000 [63] POD2 Monostrut vs vs
0/93 (0%) 4/93 (4%)
Demirag M, JHVD 2001 CC Oral coumadin on 100 SJM vs 158 1/100 (1%) CVA 0/100 (0%)
[64] POD1 and ASA Biocor vs vs
150mg/d vs Oral 1/158 (0.6%) 1/158 (0.6%)
coumadin only on
POD1
Duveau D, Eur Soc CS 3 mg/kg/d SC 349 SJM 3/349 (0.9%) 1 CVA, 2
Cardiol 1984 [65] Heparin and oral abdominal
coumadin on POD8- hemorrhage
9
Emery RW, Ann Thorac CS SC heparin and oral 1146 ATS 17/1146 (1.5%) 4/1146 (0.3%) 10/1146 (0.9%)
Surg, 2003 [66] coumadin
Fiane AE, Ann Thorac CS Oral coumadin on 997 CM 1/997 (0.1%) CVA 2/997 (0.2%) major bleeding
Surg. 1998 [67] POD 1
Iguro Y, JHVD 1999 [68] CS Oral coumadin on 473 CM 1/473 (0.2%) CVA 0/473 (0%)
POD2

Ann Thorac Surg

Lim KHH, JTCVS, 2002 RCT SC heparin and oral RCT of 234 CM 3/485 (0.6%) 1/485 (0.2%) CVA 1/485 (0.2%) hemopericardium

2006;81:770 81
[69] coumadin valves versus
251 SJM valves
Masters RG, JTCVS, CS Oral coumadin within 814 SJM and MH 2/814 (0.2%) CVA 8/814 (1.0%)
1995 [70] 72 hours postop
 2006;81:770 81
Ann Thorac Surg
Table 1. Continued
Absolute Bleeding
Study TE Mortality TE Bleeding Mortality Bleeding
Publication Type Therapy Study Description Absolute TE rate rate Description Rate Rate Description

Minakata K, JHVD 2002 CS Oral coumadin on 616 CM 1/616 (0.2%) CVA 1/616 (0.2%)
[71] POD 12
Montalescot G, Circ CC Oral coumadin with 106 mechanical 1/106 (0.9%) 1 stroke 2/106 (1.9%) 2 major bleeds
2000 [6] SC heparin (tid) valve patients
Nicoloff D, JTCVS, 1981 CS Oral coumadin and 232 SJM 1/232 (0.4%) CVA 1/232 (0.4%) Cardiac
[72] SC heparin tamponade
Nistal JF, JTCVS, 1996 CS Oral coumadin when 504 CM 3/504 (0.6%) 1 CVA, 2 14/504 (2.8%) 13 hemorrhage, 1
[73] patient started oral embolic tamponade
intake MI
Rodler SM, Ann Thorac CS SC heparin, oral 583 CM 0/583 (0%) 9/583 (1.5%)
Surg, 1997 [74] coumadin on POD4
Santini F, JHVD 2002 CS Oral coumadin on 942 CM 4/942 (0.4%) Hemorrhage
[75] POD1
Yamak B, Thorac CS Oral coumadin on 548 SJM 4/448 (0.7%) 4 CVA 37/548 (6.7)% Hemopericardium
Cardiovasc Surgeon, POD1 and cardiac
[76]
1993 tamponade
TOTAL 28/3056 (0.9%) 22/8507 (0.3%) 50/1525 (3.3%) 77/9798 (0.8%)

REVIEW OF EARLY POSTOPERATIVE ANTICOAGULATION

Intravenous Heparin
Acar J, Circ, 1996 [7] RCT IV heparin 6 hours RCT of INR 2.0 12/433 (2.7%)
postoop, oral 3.0 versus 3.0
coumadin on POD2 4.5 (mechanical
valves)
Aoyagi S, JTCVS, 1994 CS IV heparin on POD1, 908 SJM 0/908 (0%) 2/908 (0.2%)
[77] oral coumadin on
POD2
Baudet EM, JTCVS, CS IV heparin 624 hours 1112 SJM valve 3/1112 (0.3%) 3/1112 (0.3%) 3 CVA 14/1112 (1.3%) 7 surgical
1995 [78] postop then SC operations bleeding, 3
heparin, oral tamponade, 3
coumadin on day 2 intracranial

REVIEW
hemorrhage, 1
Gl bleed
Camilleri LF, CC IV heparin starting 6 134 SJM valves 5/217 (2.3%) 1/217 (0.5%) 1 valve 2/217 (0.9%) surgical bleeding
Cardiovasc Surg. 2001 hours postop, then and 86 Sorin thrombosis,

KULIK ET AL
[79] SC heparin and Bicarbon 4 CVA
oral coumadin POD2 valves
Fanikos J, Am J Cardiol CC Oral coumadin with 34 mechanical 2/34 (5.9%) 3/34 (8.8%)
2004 [8] IV heparin valve patients

773
REVIEWS
 REVIEWS

774
REVIEW OF EARLY POSTOPERATIVE ANTICOAGULATION
REVIEW KULIK ET AL
Table 1. Continued
Absolute Bleeding
Study TE Mortality TE Bleeding Mortality Bleeding
Publication Type Therapy Study Description Absolute TE rate rate Description Rate Rate Description

Laffort P, JACC, 2000 RCT IV heparin starting 6 RCT of oral 5/229 (2.2%) 0/229 (0%) 16/229 (7.0%) 1/229 (0.4%) major
[31] hours postop. then coumadin with hemorrhage
SC heparin and oral or without
coumadin on POD 2 aspirin after
SJM MVR
Thulin L, Arg Bras CS Coumadin 510 mechanical 1/510 (0.2%) NA
Cardiol 1987 [52] (intravenous and valve patients
oral) with IV
heparin
TOTAL 28/2535 (1.1%) 4/2466 (0.2%) 19/263 (7.2%) 19/2466 (0.8%)
Low Molecular Weight
Heparin
Anon, Formulary 2000 CS Oral coumadin with 37 patients, 1/37 (2.7%) 3/37 (8.1%) 2 major bleeds, 1
[58] enoxaparin mechanical and minor bleed
bioprosthetic
valves
Fanikos J, Am J Cardiol CC Oral coumadin with 29 mechanical 0/29 (0%) 3/29 (10.3%) 3 pleural
2004 [8] enoxaparin valve patients effusions
Montalescot G, Circ CC Oral coumadin with 102 mechanical 0/102 (0%) 2/102 (2.0%) 2 major bleeds
2000 [6] enoxaparin valve patients
TOTAL 1/168 (0.6%) 8/168 (4.8%)

ATS ATS valve; CC case control; CM CarboMedics valve; CS case series; CVA cerebrovascular accident; Gl gastrointestinal; INR international normalized ratio; IV
intravenous; MH Medtronic Hall valve; MI myocardial infarction; NA not available; POD postoperative day; RCT randomized controlled trial; SC subcutaneous;
SJM St. Jude Medical valve; TE thromboembolic.

Ann Thorac Surg

2006;81:770 81
Ann Thorac Surg
2006;81:770 81
Finally, as pointed out by Butchart and colleagues [17,
18], the early thromboembolic risk can also be attributed
to inconsistencies of oral anticoagulation management.
During the first six months, the thromboembolic risk is
up to seven times greater than in the after months and
years. After aortic valve replacement, the risk of throm-
boembolic events falls from 16 per 100 patient years in
the early postoperative period to 1.4 per 100 patient years
at 5 years. Similarly, after mitral valve replacement, the
risk falls from 21 per 100 patient years to 2.5 per 100
patient years [17, 18]. The propensity for early formation
of thrombus may be related to difficulties in achieving
therapeutic anticoagulation in the initial postoperative
period [18, 19]. Spot - international normalized ratio
(INR) tests of anticoagulated patients report as many as
50% of INR levels outside the therapeutic range [2], and
a recent study of prosthetic valve anticoagulation re-
ported subtherapeutic and variable anticoagulation as a
contributing factor in the majority of thromboembolic
events [22]. Variable or inadequate anticoagulation, de-
fined as an INR 25% or more below the therapeutic
range, has been demonstrated to increase the incidence
of thromboembolism 2 to 6 times, especially in the
postoperative period [2].
The early thromboembolic risk associated with the
placement of a mechanical prosthesis depends on the
complex interactions between the recipient and valve,
the intrinsic thrombogenic properties of the mechanical
valve components, and the diligent management of post-
operative anticoagulation.
ANTICOAGULATION THERAPY. The need for lifelong oral anti-
coagulation therapy in patients with mechanical pros-
thetic valves is well-recognized. In patients not receiving
long-term anticoagulation therapy, the average rate of
major thromboembolism is estimated to be 4 to 8 per 100
patient-years [23, 24]. This risk is reduced to 2.2 per 100
patient-years with antiplatelet therapy, and further re-
duced to 1 per 100 patient-years with oral anticoagulation
(warfarin). Thus, the utilization of postoperative warfarin
therapy reduces the incidence of major embolism by
approximately 75% and has become the standard of care
for all patients with mechanical prostheses [24]. In large
studies conducted on anticoagulation therapy, risk fac-
tors known to modulate the risk of thromboembolism
have included the number of valves replaced, the type of
valve implanted, mitral valve replacement (versus aortic
valve), atrial fibrillation, the presence of left atrial en-
largement or reduced cardiac output, treatment with
warfarin or other oral anticoagulants, adequacy of war-
farin therapy, and the addition of antiplatelet drugs such
as aspirin or dipyridamole [22, 24].
The optimal intensity of oral anticoagulation, defined
as the level at which the incidence of both thromboem-
bolic and bleeding complications is lowest, is still a
matter of debate. Hirsh and colleagues [25] have demon-
strated that increasing the INR, in mechanical valve
patients, above the range of 2.5 to 3.0 also increased
considerably the rate of bleeding without a reduction in
the number of thromboembolic events. The American
REVIEW KULIK ET AL 775
REVIEW OF EARLY POSTOPERATIVE ANTICOAGULATION
College of Cardiology/American Heart Association and
the American College of Chest Physicians recommended,
in their most recent guidelines, that contemporary me-
chanical valves in the aortic position be anticoagulated
with a target INR of 2.0 to 3.0, and mechanical valves in
the mitral position be anticoagulated with a target INR of
2.5 to 3.5 [26, 27]. In view of all the clinical evidence
presented, the utilization of oral anticoagulation after
placement of mechanical valves will remain the standard
of care.
ANTIPLATELET THERAPY. In recognition of the key role of
platelet activation and aggregation in thrombus forma-
tion on mechanical valve surfaces, several studies have
sought to evaluate the use of antiplatelet agents for
thromboembolic prophylaxis, either exclusively, or in
conjunction with oral anticoagulation. Ribeiro and col-
leagues [28] reported on their experience with antiplate-
let agents alone in patients with bileaflet aortic and mitral
mechanical valves. After 22 months of follow-up, there
was no thromboembolic event reported, but the valve
thrombosis rate was inappropriately high at 2.1 per 100
patient years (aortic) and 10 per 100 patient years (aortic
and mitral). In another report, Czer and colleagues [29]
found a significant difference in the rate of valve throm-
bosis between patients with no anticoagulation (18.9 per
100 patient years), antiplatelet therapy exclusively (3.2
per 100 patient years) and warfarin (1.5 per 100 patient
years). Similarly, a recent randomized trial comparing a
combination of clopidogrel and aspirin versus warfarin in
patients with mechanical aortic valves was terminated
REVIEWS
prematurely due to the increased rate of early valve
thromboses in the antiplatelet arm [30]. Therefore, in the
absence of anticoagulation, antiplatelet therapy seems to
be insufficient to prevent the development of thrombi.
Because platelet aggregation occurs despite anticoag-
ulation, the co-administration of antiplatelet agents with
anticoagulation has been advocated to reduce the fre-
quency of thromboembolic complications. Laffort and
colleagues [31] studied the addition of aspirin on postop-
erative day one with mechanical mitral valve patients
anticoagulated with intravenous heparin, followed by
oral warfarin. Patients receiving daily aspirin had a
reduction in the incidence of nonobstructive peripros-
thetic valve thrombi from 13% to 5% nine days after
surgery, and a reduced incidence of thromboembolic
events in the first 5 months, from 25% to 9% (p 0.004).
A number of studies have explored the long-term co-
administration of antiplatelet agents to chronic oral an-
ticoagulation after mechanical valve replacement, and a
recent meta-analysis by Massel and colleagues [32] dem-
onstrated that the addition of either aspirin or dipyrid-
amole to warfarin reduced the risk of thromboembolic
events by 59% (odds ratio [OR] 0.41, 95% confidence
intervals [CI] 0.29% to 0.58%) and total mortality by 51%
(OR 0.49, 95% CI 0.35% to 0.67%). Although the risk of
major bleeding was increased with antiplatelet agents
(OR 1.5, 95% CI 1.03% to 2.18%), this risk was lower with
contemporary low dose (100 mg daily) aspirin (OR 1.28).
A meta-analysis by Cappelleri and colleagues [33] inves-
 776 REVIEW KULIK ET AL
REVIEW OF EARLY POSTOPERATIVE ANTICOAGULATION
tigating the same issue demonstrated a 67% (OR 0.33,
95% CI 0.16% to 0.69%) reduction in embolism, but an
increase in major gastrointestinal hemorrhage by ap-
proximately 250% (OR 3.47, 95% CI 1.43% to 8.40%),
leading to an estimate that for every 1.6 patients who had
their stroke prevented by combination therapy, there was
an excess of one major gastrointestinal bleed. These
meta-analyses suggest that the benefits derived from the
enhanced antithrombotic potential of combined therapy
outweigh the risks [32, 33].
The American College of Cardiology/American Heart
Association, in their most recent guidelines, recom-
mended that the addition of aspirin (80 to 100 mg/day) to
warfarin be strongly considered for all patients with
mechanical valves [27]. This is particularly important for
patients who have had an embolus while on warfarin
therapy, those with known vascular disease, or those
known to be particularly susceptible to hypercoagulabil-
ity [27]. The American College of Chest Physicians, on
the other hand, recommend low-dose aspirin only for
mechanical valve patients with thromboembolic risk fac-
tors, such as atrial fibrillation, or as a strategy to lower the
target INR from 3.0 to 2.5 in patients with bileaflet
mechanical valves in the mitral position [26]. Despite the
evidence and the presence of these guidelines, a survey
of North American cardiac surgeons reported that only
21% of surgeons routinely use aspirin in conjunction with
oral anticoagulation after mechanical valve replacement
[34].
The prothrombotic milieu encountered after mechan-
REVIEWS
ical valve implantation is complex and multifactorial,
putting patients at risk for thromboembolic complica-
tions. These undesirable outcomes are largely prevented
by long-term oral anticoagulation with warfarin and
antiplatelet therapy. To date, the data in the literature
have predominantly focused on the late results of oral
anticoagulation strategies, with a paucity of data on the
effects of early postoperative anticoagulation on bleeding
and thromboembolic risk.
Early Postoperative Anticoagulation
EARLY ANTICOAGULATION AND THE RISK OF BLEEDING. A clear
relation exists between the intensity of anticoagulation in
patients with mechanical valves and the incidence of
thromboembolism, on the one hand, and of bleeding, on
the other [35]. Although intravenous heparin is fre-
quently administered after mechanical valve implanta-
tion to reduce thromboembolism, heparin has the poten-
tial to induce bleeding by inhibiting blood coagulation,
impairing platelet function [36], and increasing capillary
permeability [35]. Pericardial effusion and cardiac tam-
ponade are serious complications associated with the
early postoperative use of heparin and warfarin, and can
present days to months after surgery. While pericardial
effusion may be present in up to 64% of patients after
heart surgery, cardiac tamponade is a less frequent
event, occurring in 0.5% to 8.5% of patients [37, 38].
Defined as fluid in the pericardial space compromising
cardiac filling, cardiac tamponade is a much more serious
complication than pericardial effusion, with a mortality
Ann Thorac Surg
2006;81:770 81
rate up to 16% [37, 39]. Cardiac tamponade occurs more
often after valve replacement (11%) than coronary artery
bypass graft surgery (CABG) (2%), and is seen almost
exclusively in anticoagulated patients [38, 40]. Bleeding is
another serious complication associated with anticoagu-
lation after valve implantation. In the immediate postop-
erative period, bleeding requiring a return to the oper-
ating room is more common in valve patients than CABG
patients [41]. However, such events usually occur in the
early hours after cardiac surgery before the administra-
tion of heparin or warfarin and are therefore not usually
attributable to anticoagulation. On the other hand, bleed-
ing events, such as gastrointestinal and intracranial hem-
orrhage, usually present later in the postoperative period
and are more likely to be associated with anticoagulation.
Occurring in up to 3% of anticoagulated patients in the
initial postoperative period, the incidence of all major
bleeding events decreases to less than 3% annually in
patients on stable anticoagulation [26, 42, 43]. These
include any episodes of major internal or external bleed-
ing that cause death, hospitalization, permanent injury,
or necessitate transfusion [4]. Although upper gastroin-
testinal bleeding after cardiac surgery is a rare event
(approximately 0.5%), it is associated with a high mor-
bidity and a mortality of up to 30% [44, 45]. Important risk
factors associated with gastrointestinal bleeding after
cardiac surgery include valve replacement and anticoag-
ulation [44, 46]. In their review of postoperative gastro-
intestinal bleeding, Heikkinen and colleagues [46] re-
ported that 24% of bleeding events occurred in patients
excessively anticoagulated. Bleeding associated with the
central nervous system is also a major complication of
anticoagulation, representing 20% to 30% of all bleeding
events [2] and the most frequent cause (56%) of bleeding-
related death [43]. Systemic bleeding complications and
cardiac tamponade therefore represent serious complica-
tions associated with early postoperative anticoagulation.
The ideal early postoperative anticoagulation strategy
(with both high antithromboembolic efficacy and low
bleeding risk) has yet to be determined in a systematic
manner. Most reports on the topic are either retrospec-
tive or empirical.
EARLY POSTOPERATIVE ANTICOAGULATION STRATEGIES. Recom-
mendations for long-term anticoagulation after mechan-
ical valve replacement have been formulated by several
working groups, but there is little objective evidence
supporting strategies for the management of anticoagu-
lation in the early postoperative period. The three most
common early postoperative anticoagulation strategies
that have been reported in the literature include the
following: (1) oral warfarin starting on postoperative day
one, without the use of therapeutic heparin anticoagula-
tion; (2) intravenous unfractionated heparin beginning in
the early hours after surgery, oral warfarin starting on
postoperative day one, and the continuation of heparin
until a therapeutic INR has been achieved; (3) low mo-
lecular weight heparin (LMWH) beginning in the early
hours after surgery, oral warfarin starting on postopera-
tive day one, and the continuation of LMWH until a
Ann Thorac Surg
2006;81:770 81
therapeutic INR has been achieved. Unfractionated hep-
arin represents a heterogeneous mixture of glycosamino-
glycans that bind to antithrombin, catalyzing the inacti-
vation of thrombin and other clotting factors.
Unfractionated heparin has unpredictable pharmacoki-
netic and pharmacodynamic properties, and therefore
requires frequent laboratory assessment of the activated
partial thromboplastin time to properly monitor intrave-
nous anticoagulation therapy. The LMWH is derived
from unfractionated heparin by chemical or enzymatic
depolymerization. Compared with unfractionated hepa-
rin, LMWH has reduced antifactor IIa activity relative to
antifactor Xa activity, and lacks the nonspecific binding
affinities of unfractionated heparin. As a result, LMWH
preparations have more predictable pharmacokinetic
and pharmacodynamic properties. These properties al-
low LMWH to be administered subcutaneously once
daily without laboratory monitoring [47]. Although each
regimen has been promoted, most of the published
investigations lack data that would permit a firm conclu-
sion about the optimal anticoagulation regimen.
1. ORAL WARFARIN ALONE. With the early postoperative risks
associated with intravenous heparin, the initiation of
mechanical valve anticoagulation with oral warfarin
alone has been advocated as the safest regimen [5, 48].
This is based on the increased risk of cardiac tamponade,
which is up to five times greater in patients treated with
therapeutic intravenous heparin (1.8%) compared with
those treated with low dose heparin for deep vein throm-
bosis (DVT) prophylaxis (0.38%) [38, 49]. The safety of this
oral anticoagulation only approach was highlighted in
a recent randomized trial of different loading doses of
warfarin after valve replacement (58% mechanical valves,
42% bioprosthetic valves), without the use of intravenous
heparin. Patients randomized to receive 5 mg loading
doses achieved the therapeutic range earlier than those
randomized to 2.5 mg daily (1.98 vs 2.72 days, p 0.0001).
However, of those receiving 5 mg doses, 49.6% needed
the dose withheld for at least one day, and the gap
between the target and mean INR on day 5 was greater.
There were no bleeding complications in either group,
and despite the lack of intravenous heparin, there were
no thromboembolic complications. Therefore, a lower
loading dose of warfarin reduced the rate of excessive
anticoagulation, and the absence of intravenous heparin
was not associated with any thromboembolism [5]. Oral
anticoagulation alone may be effective at reducing early
thromboembolic and anticoagulation-related morbidity.
However, oral anticoagulation requires several days of
therapy before a therapeutic INR is achieved. Therefore,
strategies employing oral anticoagulation alone may not
be optimal in preventing the development of thrombus
in patients with mechanical valves.
2. INTRAVENOUS UNFRACTIONATED HEPARIN AND ORAL WARFARIN.
Because the risk of thromboembolism is greatest in the
early postoperative period [18], and intracardiac throm-
bus may develop soon after valve implantation, many
centers have supported the utilization of intravenous
unfractionated heparin early after surgery (within 24
REVIEW KULIK ET AL 777
REVIEW OF EARLY POSTOPERATIVE ANTICOAGULATION
hours), with oral anticoagulation starting on the first
postoperative day [1, 19, 27]. However, even with the
administration of intravenous heparin three hours after
surgery, thrombus formation can be observed by trans-
esophageal echocardiography in up to 18% of patients
after mechanical valve implantation [50].
In a swine model evaluating anticoagulation strategies
after mechanical mitral valve replacement, 69% (9/13) of
pigs died from hemopericardium within 30 days when
treated with concurrent intravenous heparin (125150
units/kg q12 hours) and oral warfarin. In this study,
heparin was discontinued when a target INR of greater
than 2.0 was achieved [51]. Thulin and Olin [52] reported
their clinical experience with anticoagulation after me-
chanical valve replacement in 510 patients. The morning
after surgery, 10 mg of warfarin sodium was adminis-
tered intravenously, followed by oral warfarin on ensu-
ing days. Anticoagulation was supplemented with intra-
venous heparin until a therapeutic INR was achieved. A
thromboembolism incidence of 0.2% within the first
three months was reported, with only 1.3% of patients
developing hemopericardium and cardiac tamponade.
The authors concluded by stating that the quality of the
anticoagulation approach was very important in the
prevention of thromboembolic and bleeding
complications.
The most recent American College of Chest Physicians
Anticoagulation Consensus Guidelines encourage the
use of intravenous unfractionated heparin until a thera-
peutic INR is achieved for mechanical valves (grade 2C
REVIEWS
recommendation), despite the lack of strong evidence
[26]. This is based on the premise that starting warfarin
therapy without simultaneous therapeutic heparin may
enhance thrombosis in situations of high thrombotic risk.
In order to avoid hypercoagulable conditions caused by
varying degrees of decay in coagulation factors, heparin
is continued until the INR is stable in the therapeutic
range [53]. Some centers favor a more aggressive antico-
agulation regimen, with the continuation of heparin
therapy for two additional days after the INR is thera-
peutic to prevent potential thromboembolic events [53,
54]. However, little controlled data are available to ascer-
tain the optimal use and the adequate degree of antico-
agulation with intravenous heparin [6]. This lack of
information and coordination has led to the establish-
ment of recommendations that are based primarily on
empirical data [26].
3. LOW MOLECULAR WEIGHT HEPARIN AND ORAL WARFARIN. The
LMWH has many potential advantages that may be
relevant for patients with mechanical heart valves. They
have a better safety profile with less thrombocytopenias
[55], less bleeding, a more predictable and rapidly
reached anticoagulant effect [56], the possibility of self-
administration without laboratory monitoring, and
shorter hospital stays and lower costs associated with
outpatient administration [57]. However, unlike unfrac-
tionated heparin, LMWH is not fully reversible by pro-
tamine, and may carry an increased bleeding risk in the
immediate postoperative period [47]. Nevertheless, the
 778 REVIEW KULIK ET AL
REVIEW OF EARLY POSTOPERATIVE ANTICOAGULATION
pharmacokinetic and biologic advantages of LMWH may
be more relevant in the postoperative period of heart
valve replacement because of the severe inflammation
and platelet and coagulation disorders related to cardio-
pulmonary bypass. This may make adequate anticoagu-
lation with unfractionated heparin more difficult [6].
Therefore, many centers have adopted the use of LMWH
within 24 hours after surgery (without the use of unfrac-
tionated heparin), and oral anticoagulation starting on
the first postoperative day [6, 8].
Several clinical trials have prospectively evaluated the
potential use of LMWH in patients after valve replace-
ment surgery. Bridging therapy with enoxaparin was
evaluated postoperatively in a case series of 37 patients
who had undergone mechanical or bioprosthetic heart
valve replacements and were stable for discharge prior to
achieving a therapeutic INR. Enoxaparin at 1 mg/kg
every 12 hours was administered until two consecutive
therapeutic INR values were reported. During the three-
month study period, bleeding related to anticoagulation
was reported in three patients (2 major, 1 minor). One
thromboembolic event was reported the day after discon-
tinuation of enoxaparin, and one death due to arrhythmia
(possibly related to anticoagulation) was reported two
weeks after discontinuation of enoxaparin. It was deter-
mined that home LMWH therapy reduced the number of
hospital days by a mean 4.6 days per patient. This
translated to health system savings of $168,300, based on
the average cost of $1,100/day for a surgical bed [58].
Montalescot and colleagues [6] performed a compara-
tive case-control study of subcutaneous unfractionated
REVIEWS
heparin (500 IU/kg/day, divided over three times a day
dosing) versus subcutaneous enoxaparin (100 anti-Xa
IU/kg every 12 hours) after heart valve replacement. Two
major bleeding events occurred in each group, and one
stroke occurred in the unfractionated heparin group.
Another case-control study of perioperative anticoagula-
tion after mechanical valve implantation compared 29
patients treated with enoxaparin (1 mg/kg every 12
hours) with 34 control patients treated with intravenous
unfractionated heparin. There were no thromboembolic
events in the LMWH group compared with two (6%) in
the unfractionated heparin group (p 0.50), and there
were three (10%) bleeding episodes (pleural effusions) in
the LMWH group compared with three (9%) bleeding
events in the unfractionated heparin group (p 1.0). The
use of LMWH reduced the length of hospital stay after
surgery (7.8 4.0 to 18.1 11.6, p 0.0001) and reduced
the cost per patient by an average of $6,864 [8]. Similarly,
an open-label randomized trial comparing the LMWH
nadroparin to intravenous unfractionated heparin after
mechanical valve implantation demonstrated that pa-
tients treated with nadroparin had a shorter hospital stay,
less coagulation tests, and easier dosing administration.
No difference in major hemorrhagic events was reported
[9]. Therefore, anticoagulation with LMWH after me-
chanical valve replacement appears to be a safe, feasible,
and effective alternative to intravenous unfractionated
heparin, and provides more predictable and rapid anti-
coagulation compared with unfractionated heparin. The
Ann Thorac Surg
2006;81:770 81
latter is an important point when considering cost-
containment issues in government funded healthcare
systems.
COMPARISON OF ANTICOAGULATION STRATEGIES. Few studies in
the literature have directly compared the risks of bleed-
ing and thromboembolism associated with different an-
ticoagulation strategies after mechanical valve replace-
ment. With limited data available, a systematic review
was conducted, searching for controlled trials and case
series involving the most common mechanical prosthetic
valves and different approaches to early postoperative
anticoagulation. Summarized in Table 1, a qualitative
comparison of the three most common early postopera-
tive anticoagulation strategies was performed with 20
studies describing the use of oral warfarin with subcuta-
neous heparin (DVT prophylactic dose) [5, 6, 59 76], 7
studies employing intravenous unfractionated heparin
and oral warfarin [7, 8, 31, 52, 7779], and 3 reports of the
use of therapeutic LMWH with oral warfarin [6, 8, 58].
Through this analysis, an estimated incidence of early
postoperative thromboembolism of 0.9%, 1.1%, and 0.6%,
respectively, was determined for each of the three anti-
coagulation strategies. Furthermore, an estimated inci-
dence of early postoperative hemorrhage of 3.3%, 7.2%,
and 4.8%, respectively, was noted. The mortality rate
secondary to thromboembolism or bleeding was similar
between oral warfarin alone (0.3% and 0.8%) compared
with intravenous unfractionated heparin with oral war-
farin (0.2% and 0.8%). These data suggest that the use of
early postoperative intravenous unfractionated heparin,
compared with the use of oral warfarin alone, yields a
substantially higher risk of bleeding without lowering the
incidence of thromboembolism. This analysis, although
simple in design, provides the most thorough up-to-date
review of the literature on early postoperative anticoag-
ulation. Nevertheless, these data cannot substitute for a
well-designed randomized controlled trial that is clearly
warranted at this time.
Through this analysis, the strategy of subcutaneous
unfractionated heparin (DVT prophylaxis) with oral cou-
madin after mechanical valve implantation yielded a
combined incidence of early postoperative thromboem-
bolism and bleeding of 4.2%. This compared with the
8.3% combined rate for a strategy of intravenous unfrac-
tionated heparin with oral coumadin after surgery. Set-
ting an level of 0.05 and of 0.10, approximately 300
patients per group, or 600 patients total would be re-
quired to adequately power a trial to detect a difference
in these two approaches during the first month after
surgery. With most surgical centers averaging 100 to 150
mechanical valve implantations per year, a multicenter
trial would be best suited to answer the question at hand.
Comment
The Need for a Nomenclature
Although the need for anticoagulant therapy in patients
with mechanical valves is not in dispute, the optimal
early postoperative anticoagulation regimen after valve
Ann Thorac Surg
2006;81:770 81
implantation has been a matter of debate. The lack of
consensus is largely dependent on achieving agreement
on standard definitions for thromboembolic events and
bleeding complications in the early postoperative period.
Certainly, some physicians might argue that the presence
of nonobstructive thrombi detected by transesophageal
echocardiography should not necessarily be labeled as a
thromboembolic event, while knowing that the risk of
thrombi expansion and embolization is present. Simi-
larly, should an episode of epistaxis be considered a
minor bleeding complication? The absence of a univer-
sally accepted nomenclature pertaining to anticoagula-
tion management may have contributed to the underre-
porting of anticoagulation-related morbidity in the
literature.
In the present setting, a comprehensive approach to
data collection related to anticoagulation would improve
clinical decision-making. While guidelines exist regard-
ing the reporting of prosthesis-related complications
during long-term follow-up after valve surgery [4], these
recommendations do not address the early postoperative
complications that may be related to anticoagulation
strategies, such as the development of a nonobstructive
prosthesis thrombus, pericardial effusion, or cardiac tam-
ponade. In order to facilitate this process, we propose the
following staging system for early postoperative throm-
boembolic and bleeding events, ranked in increasing
order of clinical severity and patient impact:
Thromboembolic Level (TEL)
TEL 1: Presence of nonobstructive thrombus in the
vicinity of the prosthetic valve as identified by
clinical imaging (echocardiography);
TEL 2: Documented evidence of peripheral limb and
visceral embolization (solid organs, coronary
thrombus);
TEL 3: Documented evidence of central nervous sys-
tem (CNS) embolization (transient ischemic
attack or cerebrovascular accident);
TEL 4: Prosthesis thrombosis requiring thrombolysis
or emergency surgery;
TEL 5: A fatal thromboembolism.
Bleeding Level (BL)
BL 1: Mild bleeding such as recurrent epistaxis or
periodontal bleeding (more than two events)
and gastrointestinal bleeds not requiring hospi-
talization or transfusions;
BL 2: Documented large pericardial effusion (1 cm
thickness on echocardiography);
BL 3: Gastrointestinal bleeding requiring hospitaliza-
tion, transfusion, endoscopic intervention or
surgery;
BL 4: Documented cardiac tamponade requiring in-
tervention;
BL 5: Documented CNS hemorrhage (cerebrovascu-
lar accident);
BL 6: Fatal bleeding.
REVIEW KULIK ET AL 779
REVIEW OF EARLY POSTOPERATIVE ANTICOAGULATION
The Need for a Trial
The initiation of an efficacious early anticoagulation pro-
tocol is important because of its potential impact on the
rate of early thromboembolic complications associated
with the prothrombotic state after mechanical valve im-
plantation. The challenge in establishing a model will be
to balance the risks of under-anticoagulation (thrombo-
embolic events) against those of over-anticoagulation
(bleeding complications) in an early postoperative setting
when the risk of bleeding may be significant. Although a
number of strategies exist, a randomized controlled trial
examining the issue of early postoperative anticoagula-
tion after mechanical valve replacement is lacking, and
the paucity of knowledge in the matter is a serious
impediment to adequate early care of patients with
mechanical prostheses. Such a study would compare two
or more early postoperative anticoagulation strategies, in
addition to routine low dose antiplatelet therapy, to
document the incidence of bleeding and thromboembolic
complications during the first month after mechanical
valve implantation. Further, it would incorporate the
proposed nomenclature described above, as well as rou-
tine postoperative echocardiography to determine the
incidence and clinical impact of prosthetic valve thrombi
and pericardial effusions in these anticoagulated pa-
tients. Clearly, a need exists for a multicenter random-
ized study comparing anticoagulation regimens after
mechanical valve replacement to determine the safest
and most efficacious approach.
REVIEWS
References
1. Chesebro JH, Adams PC, Fuster V. Antithrombotic therapy
in patients with valvular heart disease and prosthetic heart
valves. J Am Coll Cardiol 1986;8(suppl B):41B56B.
2. Edmunds LH Jr. Thrombotic and bleeding complications of
prosthetic heart valves. Ann Thorac Surg 1987;44:430 45.
3. Kearon C. Management of anticoagulation in patients who
require invasive procedures. Semin Vasc Med 2003;3:28594.
4. Edmunds LH Jr, Clark RE, Cohn LH, Grunkemeier GL,
Miller DC, Weisel RD. Guidelines for reporting morbidity
and mortality after cardiac valvular operations. J Thorac
Cardiovasc Surg 1996;112:708 11.
5. Ageno W, Turpie AG, Steidl L, et al. Comparison of a daily
fixed 2.5-mg warfarin dose with a 5-mg, international nor-
malized ratio adjusted, warfarin dose initially following
heart valve replacement. Am J Cardiol 2001;88:40 4.
6. Montalescot G, Polle V, Collet JP, et al. Low molecular
weight heparin after mechanical heart valve replacement.
Circulation 2000;101:1083 6.
7. Acar J, Iung B, Boissel JP, et al. AREVA: multicenter random-
ized comparison of low-dose versus standard-dose antico-
agulation in patients with mechanical prosthetic heart
valves. Circulation 1996;94:210712.
8. Fanikos J, Tsilimingras K, Kucher N, Rosen AB, Hieblinger
MD, Goldhaber SZ. Comparison of efficacy, safety, and cost
of low-molecular-weight heparin with continuous-infusion
unfractionated heparin for initiation of anticoagulation after
mechanical prosthetic valve implantation. Am J Cardiol
2004;93:24750.
9. Iliuta L, Candea V, Vasilescu A, Moldovan H, Gherghiceanu
DP, Macarie C. Fraxiparine versus unfractionated heparin
for the perioperative anticoagulant therapy in patients un-
dergoing mechanical prosthetic heart valve replacement.
Archives de LUnion Medicale Balkanique 2003;38:141 8.
 780 REVIEW KULIK ET AL
REVIEW OF EARLY POSTOPERATIVE ANTICOAGULATION
10. Dewanjee MK, Trastek VF, Tago M, Kaye MP. Radioisotopic
techniques for noninvasive detection of platelet deposition
in bovine-tissue mitral-valve prostheses and in vitro quan-
tification of visceral microembolism in dogs. Invest Radiol
1984;19:535 42.
11. Harker LA, Slichter SJ. Studies of platelet and fibrinogen
kinetics in patients with prosthetic heart valves. N Engl
J Med 1970;283:13025.
12. Stein B, Fuster V, Halperin JL, Chesebro JH. Antithrombotic
therapy in cardiac disease. An emerging approach based on
pathogenesis and risk. Circulation 1989;80:150113.
13. Nemerson Y, Turitto VT. The effect of flow on hemostasis
and thrombosis. Thromb Haemost 1991;66:272 6.
14. Pengo V, Peruzzi P, Baca M, et al. The optimal therapeutic
range for oral anticoagulant treatment as suggested by
fibrinopeptide A (FpA) levels in patients with heart valve
prostheses. Eur J Clin Invest 1989;19:181 4.
15. Douglas PS, Hirshfeld JW Jr, Edie RN, Harken AH, Stephen-
son LW, Edmunds LH Jr Clinical comparison of St. Jude and
porcine aortic valve prostheses. Circulation 1985;72(pt 2):
II1359.
16. Murphy DA, Levine FH, Buckley MJ, et al. Mechanical
valves: a comparative analysis of the Starr-Edwards and
Bjork-Shiley prostheses. J Thorac Cardiovasc Surg 1983;86:
746 52.
17. Butchart EG, Lewis PA, Kulatilake EN, Breckenridge IM.
Anticoagulation variability between centres: implications for
comparative prosthetic valve assessment. Eur J Cardiothorac
Surg 1988;2:72 81.
18. Butchart EG, Lewis PA, Bethel JA, Breckenridge IM. Adjust-
ing anticoagulation to prosthesis thrombogenicity and pa-
tient risk factors. Recommendations for the Medtronic Hall
valve. Circulation 1991;84(suppl):III619.
19. Heras M, Chesebro JH, Fuster V, et al. High risk of throm-
boemboli early after bioprosthetic cardiac valve replace-
ment. J Am Coll Cardiol 1995;25:11119.
20. Brais M, Braunwald NS. Acceleration of tissue ingrowth on
REVIEWS
materials implanted in the heart. Ann Thorac Surg 1976;21:
2219.
21. Bull B, Braunwald NS. Human histopathologic response to a
completely fabric-covered prosthetic heart valve. Ann Surg
1971;174:755 61.
22. Saour JN, Sieck JO, Mamo LA, Gallus AS. Trial of different
intensities of anticoagulation in patients with prosthetic
heart valves. N Engl J Med 1990;322:428 32.
23. Mok CK, Boey J, Wang R, et al. Warfarin versus dipyri-
damole-aspirin and pentoxifylline-aspirin for the prevention
of prosthetic heart valve thromboembolism: a prospective
randomized clinical trial. Circulation 1985;72:1059 63.
24. Cannegieter SC, Rosendaal FR, Briet E. Thromboembolic
and bleeding complications in patients with mechanical
heart valve prostheses. Circulation 1994;89:635 41.
25. Hirsh J, Dalen JE, Anderson DR, et al. Oral anticoagulants:
mechanism of action, clinical effectiveness, and optimal
therapeutic range. Chest 1998;114(suppl 5):445S 469S.
26. Stein PD, Alpert JS, Bussey HI, Dalen JE, Turpie AG.
Antithrombotic therapy in patients with mechanical and
biological prosthetic heart valves. Chest 2001;119(suppl 1):
220S7S.
27. Bonow RO, Carabello B, de Leon AC, Jr ??, et al. Guidelines
for the management of patients with valvular heart disease:
executive summary. A report of the American College of
Cardiology/American Heart Association Task Force on Prac-
tice Guidelines (Committee on Management of Patients with
Valvular Heart Disease). Circulation 1998;98:1949 84.
28. Ribeiro PA, Al Zaibag M, Idris M, et al. Antiplatelet drugs
and the incidence of thromboembolic complications of the
St. Jude Medical aortic prosthesis in patients with rheumatic
heart disease. J Thorac Cardiovasc Surg 1986;91:92 8.
29. Czer LS, Matloff JM, Chaux A, De Robertis M, Stewart ME,
Gray RJ. The St. Jude valve: analysis of thromboembolism,
warfarin-related hemorrhage, and survival. Am Heart J
1987;114:389 97.
Ann Thorac Surg
2006;81:770 81
30. Schlitt A, von Bardeleben RS, Ehrlich A, et al. Clopidogrel
and aspirin in the prevention of thromboembolic complica-
tions after mechanical aortic valve replacement (CAPTA).
Thromb Res 2003;109(2-3):1315.
31. Laffort P, Roudaut R, Roques X, et al. Early and long-term
(one-year) effects of the association of aspirin and oral
anticoagulant on thrombi and morbidity after replacement
of the mitral valve with the St. Jude medical prosthesis: a
clinical and transesophageal echocardiographic study. J Am
Coll Cardiol 2000;35:739 46.
32. Massel D, Little SH. Risks and benefits of adding anti-
platelet therapy to warfarin among patients with prosthetic
heart valves: a meta-analysis. J Am Coll Cardiol 2001;37:569
78.
33. Cappelleri JC, Fiore LD, Brophy MT, Deykin D, Lau J.
Efficacy and safety of combined anticoagulant and antiplate-
let therapy versus anticoagulant monotherapy after mechan-
ical heart-valve replacement: a metaanalysis. Am Heart J
1995;130(pt 1):54752.
34. Ray JG, Turpie AG. Survey of cardiac surgeons perceptions
of the addition of ASA to warfarin for patients with mechan-
ical heart valves. Can J Cardiol 1997;13:11625.
35. Levine MN, Raskob G, Landefeld S, Kearon C. Hemorrhagic
complications of anticoagulant treatment. Chest 2001;
119(suppl 1):108S21S.
36. Muriithi EW, Belcher PR, Day SP, Menys VC, Wheatley DJ.
Heparin-induced platelet dysfunction and cardiopulmonary
bypass. Ann Thorac Surg 2000;69:182732.
37. Malouf JF, Alam S, Gharzeddine W, Stefadouros MA. The
role of anticoagulation in the development of pericardial
effusion and late tamponade after cardiac surgery. Eur
Heart J 1993;14:14517.
38. Pepi M, Muratori M, Barbier P, et al. Pericardial effusion
after cardiac surgery: incidence, site, size, and haemody-
namic consequences. Br Heart J 1994;72:32731.
39. Mangi AA, Palacios IF, Torchiana DF. Catheter pericardio-
centesis for delayed tamponade after cardiac valve opera-
tion. Ann Thorac Surg 2002;73:1479 83.
40. Meurin P, Weber H, Renaud N, et al. Evolution of the
postoperative pericardial effusion after day 15: the problem
of the late tamponade. Chest 2004;125:21827.
41. Unsworth-White MJ, Herriot A, Valencia O, et al. Resternot-
omy for bleeding after cardiac operation: a marker for
increased morbidity and mortality. Ann Thorac Surg 1995;
59:664 7.
42. Landefeld CS, Goldman L. Major bleeding in outpatients
treated with warfarin: incidence and prediction by factors
known at the start of outpatient therapy. Am J Med 1989;87:
144 52.
43. van der Meer FJ, Rosendaal FR, Vandenbroucke JP, Briet E.
Bleeding complications in oral anticoagulant therapy. An
analysis of risk factors. Arch Intern Med 1993;153:1557 62.
44. van der Voort PH, Zandstra DF. Pathogenesis, risk factors,
and incidence of upper gastrointestinal bleeding after car-
diac surgery: is specific prophylaxis in routine bypass pro-
cedures needed? J Cardiothorac Vasc Anesth 2000;14:2939.
45. Sakorafas GH, Tsiotos GG. Intra-abdominal complications
after cardiac surgery. Eur J Surg 1999;165:820 7.
46. Heikkinen LO, Ala-Kulju KV. Abdominal complications fol-
lowing cardiopulmonary bypass in open-heart surgery.
Scand J Thorac Cardiovasc Surg 1987;21:17.
47. Hirsh J, Raschke R. Heparin and low-molecular-weight
heparin: the Seventh ACCP Conference on Antithrombotic
and Thrombolytic Therapy. Chest 2004;126(suppl 3):188S
203S.
48. Chesebro JH, Fuster V. Optimal antithrombotic therapy for
mechanical prosthetic heart valves. Circulation 1996;94:
2055 6.
49. Wilhelm M, Carrel T, Jenni R, Turina M. Incidence, clinical
aspects and therapy of late pericardial tamponade following
heart surgery. Swiss Surg 1996;3:92 6.
50. Malergue MC, Temkine J, Slama M, et al. Value of early
systematic postoperative transesophageal echocardiography
Ann Thorac Surg
2006;81:770 81
in mitral valve replacements. A prospective study of 50
patients. Arch Mal Coeur Vaiss 1992;85:1299 304.
51. Grehan JF, Hilbert SL, Ferrans VJ, Droel JS, Salerno CT,
Bianco RW. Development and evaluation of a swine model
to assess the preclinical safety of mechanical heart valves.
J Heart Valve Dis 2000;9:710 9; discussion 719-20.
52. Thulin LI, Olin CL. Initiation and long-term anticoagulation
after heart valve replacements. Arq Bras Cardiol 1987;49:
265 8.
53. Ginsberg JS. Management of venous thromboembolism.
N Engl J Med 1996;335:1816 28.
54. British Society for Haematology annual scientific meeting.
Glasgow, United Kingdom, 27-30 April 1998. Proceedings
and abstracts. Br J Haematol 1998;101(suppl 1):1115.
55. Warkentin TE, Levine MN, Hirsh J, et al. Heparin-induced
thrombocytopenia in patients treated with low-molecular-
weight heparin or unfractionated heparin. N Engl J Med
1995;332:1330 5.
56. Weitz JI. Low-molecular-weight heparins. N Engl J Med
1997;337:688 98.
57. Anderson DR, OBrien BJ, Levine MN, Roberts R, Wells PS,
Hirsh J. Efficacy and cost of low-molecular-weight heparin
compared with standard heparin for the prevention of deep
vein thrombosis after total hip arthroplasty. Ann Intern Med
1993;119:110512.
58. Anon. Home LMWH bridge therapy in cardiac replacement:
safe, effective, and cost-saving. Formulary 2000;35:990 1.
59. Akins CW. Long-term results with the Medtronic-Hall val-
vular prosthesis. Ann Thorac Surg 1996;61:806 13.
60. Anttila V, Heikkinen J, Biancari F, et al. A retrospective
comparative study of aortic valve replacement with St. Jude
medical and medtronic-hall prostheses: a 20-year follow-up
study. Scand Cardiovasc J 2002;36:539.
61. Bernal JM, Rabasa JM, Gutierrez-Garcia F, Morales C, Nistal
JF, Revuelta JM. The CarboMedics valve: experience with
1,049 implants. Ann Thorac Surg 1998;65:137 43.
62. Dalrymple-Hay MJ, Pearce RK, Dawkins S, et al. Mid-term
results with 1,503 CarboMedics mechanical valve implants.
J Heart Valve Dis 2000;9(3):389 95.
63. de la Fuente A, Sanchez R, Romero J, et al. CarboMedics and
Monostrut valves: clinical and hemodynamic outcomes in a
randomized study. J Heart Valve Dis 2000;9:3037.
64. Demirag M, Kirali K, Omeroglu SN, et al. Mechanical versus
biological valve prosthesis in the mitral position: a 10-year
follow up of St. Jude Medical and Biocor valves. J Heart
Valve Dis 2001;10:78 83.
65. Duveau D, Michaud JL, Despins P, et al. Mitral valve
replacement with St Jude Medical prosthesis. Incidence of
thromboembolic events in 349 patients. Eur Heart J 1984;
5(suppl D):49 52.
REVIEW KULIK ET AL 781
REVIEW OF EARLY POSTOPERATIVE ANTICOAGULATION
66. Emery RW, Van Nooten GJ, Tesar PJ. The initial experience
with the ATS Medical mechanical cardiac valve prosthesis.
Ann Thorac Surg 2003;75:444 52.
67. Fiane AE, Geiran OR, Svennevig JL. Up to eight years
follow-up of 997 patients receiving the CarboMedics pros-
thetic heart valve. Ann Thorac Surg 1998;66:443 8.
68. Iguro Y, Moriyama Y, Yamaoka A, et al. Clinical experience
of 473 patients with the omnicarbon prosthetic heart valve.
J Heart Valve Dis 1999;8:674 9.
69. Lim KH, Caputo M, Ascione R, et al. Prospective random-
ized comparison of CarboMedics and St Jude Medical bileaf-
let mechanical heart valve prostheses: an interim report.
J Thorac Cardiovasc Surg 2002;123:2132.
70. Masters RG, Pipe AL, Walley VM, Keon WJ. Comparative
results with the St. Jude Medical and Medtronic Hall me-
chanical valves. J Thorac Cardiovasc Surg 1995;110:66371.
71. Minakata K, Wu Y, Zerr KJ, et al. Clinical evaluation of the
carbomedics prosthesis: experience at providence health
system in Portland. J Heart Valve Dis 2002;11:844 50.
72. Nicoloff DM, Emery RW, Arom KV, et al. Clinical and
hemodynamic results with the St. Jude Medical cardiac valve
prosthesis. A three-year experience. J Thorac Cardiovasc
Surg 1981;82:674 83.
73. Nistal JF, Hurle A, Revuelta JM, Gandarillas M. Clinical
experience with the CarboMedics valve: early results with a
new bileaflet mechanical prosthesis. J Thorac Cardiovasc
Surg 1996;112:59 68.
74. Rodler SM, Moritz A, Schreiner W, End A, Dubsky P,
Wolner E. Five-year follow-up after heart valve replacement
with the CarboMedics bileaflet prosthesis. Ann Thorac Surg
1997;63:1018 25.
75. Santini F, Casali G, Viscardi F, et al. The CarboMedics
prosthetic heart valve: experience with 1,084 implants.
J Heart Valve Dis 2002;11:121 6; discussion 27.
76. Yamak B, Karagoz HY, Zorlutuna Y, Eralp A, Tasdemir O,
Bayazit K. Low-dose anticoagulant management of patients
REVIEWS
with St. Jude Medical mechanical valve prostheses. Thorac
Cardiovasc Surg 1993;41:38 42.
77. Aoyagi S, Oryoji A, Nishi Y, Tanaka K, Kosuga K, Oishi K.
Long-term results of valve replacement with the St. Jude
Medical valve. J Thorac Cardiovasc Surg 1994;108:10219.
78. Baudet EM, Puel V, McBride JT, et al. Long-term results of
valve replacement with the St. Jude Medical prosthesis.
J Thorac Cardiovasc Surg 1995;109:858 70.
79. Camilleri LF, Bailly P, Legault BJ, Miguel B, DAgrosa-
Boiteux MC, de Riberolles CM. Mitral and mitro-aortic valve
replacement with Sorin Bicarbon valves compared with St.
Jude medical valves. Cardiovasc Surg 2001;9:272 80.