Journal of the American College of Cardiology
© 2004 by the American College of Cardiology Foundation
Published by Elsevier Inc.
Vol. 43, No. 7, 2004
ISSN 0735-1097/04/$30.00
doi:10.1016/j.jacc.2003.09.064
Valvular Heart Disease
Clinical Significance of Early Thrombosis
After Prosthetic Mitral Valve Replacement
A Postoperative Monocentric Study of 680 Patients
Guillaume Laplace, MD, Stéphane Lafitte, MD, PHD, Jean-Noèl Labèque, MD, Jean-Marie Perron, MD,
Eugène Baudet, MD, Claude Deville, MD, Xavier Roques, MD, Raymond Roudaut, MD, FESC
Chu de Bordeaux, Pessac, France
The aim of this study was to evaluate the incidence of early thrombosis, its prognostic
significance, and the therapeutic implications.
BACKGROUND Transesophageal echocardiography (TEE) is the method of choice for detecting symptomless
early postoperative thrombosis of prosthetic valves. However, the clinical significance is not
yet known.
METHODS
Between June 1994 and December 2000, 680 consecutive patients underwent TEE on day 9
after mechanical mitral valve replacement, to search for early thrombosis. Initially, end points
were the in-hospital outcome and treatment. Patients were also evaluated 34 ⫾ 22 months
after surgery.
RESULTS
Sixty-four early thrombi were detected (9.4%). Two early obstructive were treated by
redo-surgery. Sixty-two nonobstructive benefited from medical treatment. The patients were
allocated into two groups as a function of the maximum size of thrombus: ⬍5 mm in 29
patients (group A) and ⱖ5 mm in 35 (group B). During early follow-up, we observed one
complicated course in group A and eight in group B. In the long-term survey, three
complications were noted in group A and 11 in group B. Incidence of early (p ⫽ 0.027) and
long-term (p ⫽ 0.04) complications were significantly different in the two subsets.
CONCLUSIONS This study confirms the incidence of early thrombi after mechanical mitral valve replacement
detected by TEE. A small (⬍5 mm) nonobstructive thrombus seems benign, and our
experience argues in favor of medical treatment. Prognosis appears more serious for large
thrombi. Medically aggressive therapy and further surgery should be considered in cases of
obstructive thrombosis or large and mobile nonobstructive thrombosis. (J Am Coll Cardiol
2004;43:1283–90) © 2004 by the American College of Cardiology Foundation
OBJECTIVES
A number of recent studies with transesophageal echocardiography (TEE) have identified a high incidence of nonobstructive thrombi (NOT) in the early postoperative period after mechanical mitral valve replacement (MMVR)
(1–5). Such phenomena are known to increase morbidity
and mortality and are often the consequence of inadequate
anticoagulant therapy in the highly unstable period after
surgery (3,6,7). Transesophageal echocardiography is currently recognized as the method of choice for detecting
abnormal echoes, especially those from NOT.
The purpose of our study was to define the incidence of
nonobstructive and obstructive thrombosis after implantation of bileaflet mitral prostheses, the optimal management,
and the short- and long-term complications of these
thrombi.
METHODS
Population. A total of 680 consecutive patients who benefited from MMVR with a bileaflet prosthesis (Saint Jude
From the Hôpital Cardiologique du Haut-Lévêque, Chu de Bordeaux, Pessac,
France. Robert A. O’Rourke, MD, acted as the guest editor for this paper.
Manuscript received March 13, 2003; revised manuscript received August 19, 2003,
accepted September 23, 2003.
Downloaded From: https://content.onlinejacc.org/ on 09/29/2016
Medical, MIRA, ATS, SORIN) from June 1994 to December 2000 in our institution were included in this
prospective, monocentric study. All patients underwent
systematic, early TEE on day 9 after surgery. The postoperative anticoagulant treatment consisted of intravenous
heparin started 6 h after implantation and adjusted from the
activated partial thromboplastin time (aPTT). This was
followed at the 24th h by subcutaneous heparin three times
a day, which was prolonged until the target international
normalized ratio (INR ⫽ 2.5 to 3.5) was obtained with oral
anticoagulant therapy (fluindione), started at day 2. The first
219 patients were randomized in a prospective study to
determine the effect of a low dose of aspirin (200 mg) in the
postoperative period (8). The other 461 patients did not
receive any antiplatelet therapy.
Echocardiography. Transesophageal echocardiography
was performed in the early postoperative period in all
patients, using a Hewlett-Packard Sonos 2500 or 5000
(Hewlett-Packard Imaging System, Andover, Massachusetts) and Acuson Sequoia (Acuson, Mountain View, California) connected to a multiplane 5-7 Mhz probe. All TEE
were recorded on videotapes and reexamined twice. We
carefully examined the atrial and ventricular side of the
mitral prosthesis, looking for leaflet dysfunction, mitral
1284
Laplace et al.
Early Thrombi After Mitral Valve Surgery
JACC Vol. 43, No. 7, 2004
April 7, 2004:1283–90
Abbreviations and Acronyms
aPTT ⫽ activated partial thromboplastin time
INR
⫽ international normalized ratio
MMVR ⫽ mechanical mitral valve replacement
NOT ⫽ nonobstructive thrombus
OT
⫽ obstructive thrombus
SEC
⫽ spontaneous echocardiographic contrast
TEE
⫽ transesophageal echocardiography
TIA
⫽ transient ischemic accident
TTE
⫽ transthoracic echocardiography
regurgitation, spontaneous echocardiography contrast
(SEC), strands, and thrombi. A strand was defined as a
linear, mobile, and thin (⬍1 mm) echo attached to the atrial
side of the prosthesis. A thrombus was defined as an
abnormal well circumscribed mobile or immobile mass
usually found in the vicinity of the mitral valve annulus. All
characteristics of the thrombi were analyzed to specify size,
form, localization, and mobility.
In accordance with the report of Guéret et al. (3), thrombi
were defined as small when their maximal length was ⬍5 mm
(group A) and large when the length was 5 mm (group B).
Therapeutic approach and survey of thrombus. The therapy of the thrombi was left to the discretion of each surgical
team and was analyzed retrospectively. Clinical follow-up,
according to published recommendations (9), included total
and cardiac mortality, along with morbid events: symptomatic embolic event (stroke, transient ischemic attack [TIA]
of less than 24 h duration, peripheral embolism), prosthetic
obstruction, and treatment complications such as hemorrhage. Transesophageal echocardiography monitoring was
left to the discretion of the surgical and medical teams and
was carried out in about half the cohort.
Follow-up. Early clinical course (one month) was studied
during the in-hospital period. Long-term follow-up was
conducted by telephone or letter contact to the consulting
physicians at 34 ⫾ 22 months since implantation.
Statistical analysis. All data are expressed as mean ⫾ 1 SD.
Qualitative values were compared by Fischer exact test. Unpaired Student t test was used to the quantitative variables
analysis. The p value indicating statistical significance was set
at 0.05.
RESULTS
Early systematic postoperative TEE (day 9) was performed
in 680 patients from June 1994 to December 2000, revealing
in 64 patients the presence of one or more abnormal masses
on the prosthesis interpreted as local thrombus (9.4%).
The cohort comprised 33 women and 31 men with a
mean age of 64 ⫾ 9 years (range, 39 to 82 years) (Table 1).
At the time of the TEE, all patients were treated by oral
anticoagulation (fluindione), and 31 of them (47%) were
still treated by subcutaneous or intravenous heparin when
INR targets had not been reached.
The bileaflet valve was a St. Jude Medical prosthesis (St.
Downloaded From: https://content.onlinejacc.org/ on 09/29/2016
Figure 1. Transesophageal echocardiography showing two round nonobstructive thrombi on the atrial side of the sewing ring of the mitral
prosthesis.
Paul, Minnesota) in 38 cases, ATS (AB Medical, Minneapolis, Minnesota) for 14 patients, SORIN (Saluggia, Italy),
and MIRA (Edwards Lifesciences, Mairepas, France) in six
cases.
On day 9, only two patients were symptomatic: one
presented with nonfatal stroke and the other with congestive heart failure (New York Heart Association class 4).
However, these two patients presented a NOT on the
prosthesis. Sixty-two others patients were either asymptomatic or had minor aspecific symptoms.
In two asymptomatic patients, transthoracic echocardiography (TTE) and TEE detected early OT, measuring
more than 10 mm, and both were treated by redo-surgery
(one mechanical valve, one bioprosthetic valve).
In the other 62 patients, transthoracic examination suggested normal function of the bileaflet prosthesis (no significant dysfunction of the poppet, no increase in transvalvular mean gradient). In many of them, TEE showed
strands or spontaneous contrast echoes, but these phenomena were not considered as pathologic in our study.
Echographic characteristics of thrombi. SIZE. Maximum
diameter of thrombus was less than 5 mm in 29 cases (45.3%)
(Figs. 1 and 2). Thrombus size was from 5 to 10 mm in 28
patients (43.7%), and thrombi of more than 10 mm were
detected in seven patients (11%) including the two initial
obstructive ones.
LOCALIZATION. Sixty-two thrombi were attached to the
atrial side of the prosthesis (on the sewing ring or on the
poppet), but we also found four thrombi on the ventricular
side of the annulus.
FORM. The abnormal mass was localized and round in 40
cases, and circumferential in 24 patients.
Thirty-six thrombi (56%) were sessile and immobile; 28 (44%) were pedunculated and slightly mobile.
MOBILITY.
JACC Vol. 43, No. 7, 2004
April 7, 2004:1283–90
Figure 2. Same transesophageal echocardiography as in Figure 1 but with
a perpendicular view. Thrombus finally appears annular and immobile.
Two predisposing factors of early
thrombosis were analyzed: atrial fibrillation was found in 26
patients of 64 (41%) before or immediately after surgery.
Sixteen patients (25%) were operated with an impaired left
ventricle ejection fraction (⬍50%), and the association of
both factors was observed in eight patients (12%). Three
patients were implanted with a pacemaker in the postoperative period, and the anticoagulant therapy was, therefore,
discontinued for a few hours.
PREDISPOSING FACTORS.
THERAPEUTIC APPROACHES. Immediate therapy after the
discovery of early postoperative thrombus was left to the
discretion of each surgical team: the decision for reintervention was based on clinical and echographic data. The
two early obstructive thrombi were immediately reoperated because of the major risk of complication; medical
treatment was readjusted in all of the remaining cases of
early NOT. Intravenous or subcutaneous heparin was
reinitiated in 33 patients (53%) monitored by the aPTT.
A simple readjustment and optimization of the oral
anticoagulation (vitamin K antagonist therapy) with a
target INR ⬎3 was proposed for 19 patients (31%).
Aspirin was added in 13 cases and nonsteroid antiinflammatory therapy in seven patients.
Clinical course of patients with early thrombi. Immediate
(⬍1 month) and long-term (⫽1 year) outcomes were
analyzed in the cohort of 64 patients exhibiting an early
thrombus after MVR (Table 1). The mean duration of
follow-up was 34 ⫾ 22 months and was obtained for the
majority of patients, with only two lost to follow-up during
the first year.
Early course (one month). In group A (small thrombus,
n ⫽ 29), we only observed one early (first month of
follow-up) complicated course (3.4%): one patient had a
TIA (defined as a neurologic syndrome of less than 24 h
duration) and died of heart failure at postoperative day 15
(Patient 8).
Among the 35 patients of group B (large thrombus, n ⫽
Downloaded From: https://content.onlinejacc.org/ on 09/29/2016
Laplace et al.
Early Thrombi After Mitral Valve Surgery
1285
35), the early course seemed to be more serious. Two early
obstructive thrombi were immediately reoperated, which is
recognized as a complicated course (Patient 42 and Patient
63); in the other 33 patients with large NOT, six presented
severe events: three strokes, defined as a sudden focal brain
dysfunction lasting more than 24 h (Patients 2, 7, and 53),
and two secondary enlargements, finally becoming obstructive in the first month despite anticoagulant therapy, and
requiring redo-surgery (Patients 21 and 24); one of these
patient died of multiple organ failure during the first month.
In this cohort, we also observed another early fatal event: a
patient with two NOT measuring 5 mm died as a result of
gastrointestinal bleeding (Patient 13). In total, eight of the
35 patients with large thrombi presented an early (first
month) complicated course (22%).
Long-term course (more than one year). Over the whole
follow-up period (mean duration: 34 ⫾ 22 months), we
observed two other complications in group A: two late TIA
(Patient 19 in the second and Patient 40 in the fourth
month). In group B, a trend for more serious complications
was noted: two late nonfatal strokes (Patients 9 and 52) and
three other deaths in this cohort. One patient with an early
stroke died six months later (Patient 2), and another of heart
failure at month 6 (Patient 41). One of the two patients
with early OT, despite redo-surgery, presented a fatal course
12 weeks after MMVR (Patient 42).
It is of note that a patient with an early stroke had
another neurological ischemic event five years later. In this
patient, a small thrombus was seen on the valve, but this
event was not recorded as a complication because of the
disappearance of the first thrombus on TEE examination
five months after implantation. Total mortality was estimated at 9.4% (six patients) in the long-term survey (three
early and three late deaths), and five of these six patients
were in group B.
Finally, in group A, the follow-up revealed a complicated
course for 3.4% of patients in the first month and for 10%
over the total duration of the study. In group B, complications occurred in 22% of patients in the early period and in
31% over the long-term survey. Complications were significantly more frequent during the early (p ⫽ 0.027) and
long-term (p ⫽ 0.04) follow-up periods in the group of
patients with large thrombi (group B).
From another viewpoint, we could split this large cohort
into two different subsets: the first of 54 patients with an
early (⬍1 month) uneventful course (in this group, mean
size of thrombus ⫽ 5.56 ⫾ 3.3 mm). The second group
comprised 10 patients presenting an early complication after
diagnosis of early thrombosis, whose mean lengths (8.95 ⫾
5.3 mm) were significantly higher than those of the first
group (p ⫽ 0.0009). A significant difference was also noted
between these two subsets in the long-term follow-up. We
concluded that the early and long-term course appeared to
depend on the initial size of the thrombus with a higher
incidence of events (principally embolic accidents) for patients with large, often asymptomatic thrombi. These results
1286
Table 1. Summary of 64 Cases of Early Postoperative Thrombosis After MMVR: Demographic, Echographic, and Clinical Data
M
69
SJM
AF/imp LVEF
9
2
F
75
SJM
AF/imp LVEF
3
4
F
F
82
70
SJM
SJM
5
6
7
F
F
F
61
61
66
SJM
SJM
SJM
8
9
10
11
12
13
F
M
M
F
F
M
70
67
55
40
66
67
SJM
SJM
SJM
SJM
SJM
SJM
14
15
16
17
18
19
F
M
F
M
M
F
74
71
70
66
55
67
SJM
SJM
SJM
SJM
SJM
SJM
20
21
M
F
63
71
SJM
SJM
Imp LVEF/PM
22
23
24
F
M
F
68
60
73
SJM
SJM
SJM
Imp LVEF
25
26
27
28
29
30
31
32
33
34
F
M
M
M
M
M
M
M
M
F
58
59
54
54
75
57
60
68
67
67
SJM
SJM
SJM
SJM
Sorin
Sorin
SJM
Sorin
ATS
Sorin
AF
1
AF
Stroke at day 7
Imp LVEF
AF
AF/imp LVEF
AF/imp LVEF
AF
AF
Imp LVEF
PM
AF
AF
Imp LVEF
Obstruction
Treatment
Follow-Up
(Month)
68
7
Hep-OAC
readjusted
OAC-asp
5
5
Hep-OAC
Hep-OAC
66
65
3
3
6
Hep-OAC
Hep-OAC
OAC
62
54
84
3
6
3
3
4.5
5
OAC
OAC-asp-NSAI
OAC-asp
OAC-asp
OAC
OAC
0.5
69
70
60
69
1
4
2
2
4
4
3
OAC
Hep-OAC
Hep-OAC
Hep-OAC-asp
Hep-OAC
Hep-OAC-asp
53
76
52
6
52
53
Hep-OAC
Hep-OAC-redosurgery
Hep-OAC
OAC
Hep-redo-surgery
49
1
20
14
Heart failure
AF
Size of
Thrombus
(mm)
10
7
10
4
3
3
8
11
4
4
4
12
5
Secondary Obs
Secondary Obs
OAC
Hep-OAC-asp
OAC
Hep-OAC
Hep-OAC-asp
OAC
Hep-OAC
Hep-OAC-asp
OAC
6
52
41
40
64
37
12
20
29
Lost
28
27
27
28
Evolution
Short-Term
Long-Term
TEE Survey
Dis at 6 months
Stroke
Stroke
Death at 6
months
TIA 5 years
later
Per at 6 months
Dis at 7 months
Dis at 6 months
Dis at 6 months
Dis at day 7
TIA-death
Stroke
Per at 6 months
Dis at 5 months
Dis at 5 months
Dis at 6 months
Death by
GIB
TIA at 2
months
Dis at 5 months
Dis at 5 months
Dis at 5 months
Dis at 6 months
Dis at 6 months
Per at 5 months
Obs death
Dis at 5 months
Obs heart
failure
Dis at 5 months
Dis at 5 months
Per at 1 month
Dis at 1 month
Per at 2 months
Dis at 5 months
Continued on next page
Downloaded From: https://content.onlinejacc.org/ on 09/29/2016
JACC Vol. 43, No. 7, 2004
April 7, 2004:1283–90
Risk Factor
Gender
Initial
Symptoms
Laplace et al.
Early Thrombi After Mitral Valve Surgery
Age
Type
of
Valve
Pt.
No.
Gender
Age
Type
of
Valve
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
F
M
M
F
M
F
M
M
F
F
F
M
M
F
F
M
F
F
M
58
50
75
54
73
65
72
72
58
59
77
39
51
68
65
47
60
60
73
Sorin
Sorin
Sorin
SJM
ATS
Mira
ATS
ATS
SJM
Mira
Mira
ATS
SJM
ATS
Mira
SJM
ATS
ATS
ATS
54
55
56
57
58
F
M
M
F
M
49
62
70
45
69
ATS
SJM
ATS
Mira
ATS
59
60
61
62
63
64
M
F
M
F
F
F
66
77
59
71
71
52
SJM
SJM
ATS
Mira
ATS
SJM
Risk Factor
AF/imp LVEF
AF
AF
AF
AF
AF
Imp LVEF/PM
AF/imp LVEF
AF
AF/imp LVEF
Imp LVEF
AF
Imp LVEF
AF
AF
AF/imp LVEF
AF
Initial
Symptoms
Size of
Thrombus
(mm)
8
7
4
6
4
2
6
11
5
3
5
4
6
10
5
10
7
12
10
Obstruction
Treatment
Follow-Up
(Month)
Hep-OAC
Hep-OAC
Hep-OAC
OAC
Hep-OAC-asp
OAC
OAC-asp
Redo-surgery
OAC
OAC
Hep-OAC
OAC
OAC-asp-NSAI
OAC
OAC
Hep-OAC
OAC
Hep-OAC-asp
Hep-OAC
25
52
26
Lost
25
24
6
4
48
45
33
30
42
12
14
14
12
16
15
6
11
2
3
5
Hep-OAC
Hep-OAC-NSAI
OAC-NSAI
Hep-OAC-NSAI
Hep-OAC
11
29
28
28
12
2
3
2
6
20
8
OAC-asp
Hep-OAC
OAC
Hep-OAC
Redo-surgery
Hep-OAC
12
12
27
24
1
40
Initial Obs
Initial Obs
Evolution
Short-Term
Long-Term
TEE Survey
Dis at 1 month
Dis at 1 month
TIA at 4 months
Death at 6 months
Death at 4 months
Dis at 1 year
Stroke at 5 month
Stroke at 1st
month
Dis at 7 months
Dis at 3 months
Dis at 1 month
Decrease at 1
month
Dis at 1 month
Decrease at day 8
AF ⫽ atrial fibrillation; asp ⫽ aspirin; Dis ⫽ disappearance; F ⫽ female; GIB ⫽ gastrointestinal bleeding; Hep ⫽ heparin; imp LVEF ⫽ impaired left ventricle ejection fraction; M ⫽ male; MMVR ⫽ mechanical mitral valve replacement;
NSAI ⫽ nonsteroid anti-inflammatory; Obs ⫽ obstruction; OAC ⫽ oral anticoagulation; Per ⫽ persistence; PM ⫽ pacemaker implantation; Pt. No. ⫽ patient number; TEE ⫽ transesophageal echocardiography; TIA ⫽ transient ischemic
attack.
Laplace et al.
Early Thrombi After Mitral Valve Surgery
Pt.
No.
JACC Vol. 43, No. 7, 2004
April 7, 2004:1283–90
Table 1. Continued
1287
Downloaded From: https://content.onlinejacc.org/ on 09/29/2016
1288
Laplace et al.
Early Thrombi After Mitral Valve Surgery
JACC Vol. 43, No. 7, 2004
April 7, 2004:1283–90
Table 2. Early (ⱕ1 Month) and Long-Term Evolution of
Thrombosis as a Function of the Size
Initial Size
<5 mm
>5 mm
p Value
Early complicated course
Long-term complicated course
1 (3.3%)
3 (10%)
8 (22%)
11 (31%)
0.027
0.04
are listed in Table 2 and Figure 3 and would indicate a
revised therapeutic approach for patients with large
thrombi.
DISCUSSION
There are numerous reports of a high incidence of postoperative NOT after MMVR (1–5,7). Our study represents
the largest population to date of patients benefiting from
systematic TEE after surgery. We showed that about 10%
of these patients presented an abnormal mass usually adhering to the atrial face of the prosthesis, considered as a
thrombus. It should be noted that all the patients except two
examined in this cohort of 680 presented few or no
symptoms on the day of early TEE.
Systematic TEE after MMVR. Diagnosis of mechanical
valve thrombosis is usually based on TTE and cinefluoroscopy; on TTE, thrombosis can be detected from an increase
in the transvalvular mean pressure gradient measured by
Doppler. A reduction in mobility or even immobility of one
of the poppets may also be evidenced. However, an increase
in this gradient is commonly seen immediately after surgery
due to the altered hemodynamic conditions, the presence of
anemia or tachycardia, even when the prosthesis works
perfectly. On the other hand, normal transthoracic echocardiogram examination and fluoroscopy may not exclude
NOT. Thus, TTE and cinefluoroscopy seem to be attractive
alternatives (5). However, TEE is recognized as having
superior sensitivity and specificity for the diagnosis of mitral
prosthetic thrombi because of echocardiography artefacts in
the left atrium on TTE examination (10 –14). Many studies
have demonstrated the interest of TEE after mechanical
MVR for the detection of thrombi (7–9). In our experience
on 680 patients, we observed 64 thrombi (9.4%), most of
them nonobstructive. Our patients (except two) presented
few symptoms indicative of thrombotic phenomena, and
their valves were considered to be functioning normally
based on the postoperative Doppler parameters from systematic TTE. This incidence had been demonstrated in
many prospective studies of systematic TEE after MVR.
However, in these asymptomatic patients, we observed a
higher incidence of early and late complications, such as
systemic embolic events (TIA, stroke) or obstruction with
heart failure. We found a relationship between the initial
size of the thrombus and the incidence of complications.
These observations would indicate the value for a thorough
search for early thrombosis after MMVR, as a guide to
postoperative management. We, therefore, recommend, as
mentioned in French guidelines for the management of
patients with valvular heart diseases (15), systematic postoperative TEE after MMVR.
Figure 3. Early and long-term evolution of thrombi in the two groups of patients. G.I.B. ⫽ gastrointestinal bleeding; O.T. ⫽ obstructive thrombus; T.I.A.
⫽ transient ischemic accident.
Downloaded From: https://content.onlinejacc.org/ on 09/29/2016
JACC Vol. 43, No. 7, 2004
April 7, 2004:1283–90
Predisposing factors for thrombosis. Many authors have
looked for hemodynamic or echographic factors predictive
of early thrombi after mitral surgery. Dadez et al. (2)
showed that SEC in the left atrium as detected by TEE was
associated with abnormal small echoes on the atrial side of
the prosthesis, which were responsible for the increase in
embolic events. This notion was supported by Bonnefoy et
al. (4) in a study of systematic TEE 24 h after MVR. Özkan
(16) described three independent predictors: SEC, atrial
fibrillation, and preoperative thrombus. However, Malergue
et al. (1) failed to evidence any statistical difference for
incidence of atrial fibrillation, SEC, or left ventricle dysfunction. Guéret (3) concluded that early thrombosis after
mitral prosthesis was associated with unpaired systolic
fractional shortening and atrial parameters. According to
Iung (6), small abnormal echos after MMVR are associated
with conditions predisposing to SEC with high dose protamine administration during surgery compounded by inadequate postoperative anticoagulation.
In a previous study in our center (7), the low level of
anticoagulation in the immediate postoperative period
emerged as the principal cause of early thrombosis. Indeed,
the formation of a thrombus can be explained by an
imbalance in the parameters of Virchow (17): cardiac
surgery, especially mechanical MVR, and extracorporeal
circulation damages native tissues, while artificial surfaces
induce variations in local rheology with stagnation of blood
and reduction in flow velocity. A hypercoagulable state is
often observed with platelet and leucocyte activation, with
coagulation factors and fibrinogen maintaining a vicious
circle. All these parameters are affected by poor anticoagulation, which is thought to represent the most important
risk factor (3,6,7).
Therapeutic implications. Management of thrombus after
mechanical MVR is still matter of debate (18). For obstructive thrombi, the major risk of complications indicates an
early aggressive approach according to American College of
Cardiology/American Heart Association guidelines (19).
Fibrinolysis is contraindicated in the first 14 postoperative
days due to risk of major bleeding (hemopericardium), but
may be indicated at later times (18,20 –26). Guéret (3)
recommends systematic surgical treatment for obstructive or
large and threatening thrombi, or failure of medical therapy
(complications or thrombi growing under anticoagulant
therapy). Although redo-surgery carries a major risk of
mortality (7), we applied this strategy to the two initial and
two secondary OT, with long-term success in three of these
patients.
The problem of management of NOT is more acute (24).
The evolution of these thrombi seems to depend on the
initial size as found in the present study. Several authors
have shown the importance of anticoagulation in the immediate postoperative period. We also recommend the
reinitialization of intravenous continue heparin monitored
by aPTT until correct adjustment of oral anticoagulation.
The value of aspirin is still unclear. Laffort et al. (8)
Downloaded From: https://content.onlinejacc.org/ on 09/29/2016
Laplace et al.
Early Thrombi After Mitral Valve Surgery
1289
showed that aspirin added to conventional oral anticoagulation decreased the incidence of postoperative thrombi, but
long-term prescription increased the incidence of major
bleeding (gastrointestinal hemorrhage). We propose an
association of anticoagulation and aspirin with gastric protection for large and mobile NOT; this group of patients
will probably require TEE survey until disappearance of the
thrombi. However, the incidence of potential thromboembolic events in the subset with large thrombi might justify a
more aggressive approach.
Study limitations. The principal limitation of the present
study was the absence of randomization in the therapeutic
strategies. Each surgical team chose how to manage their
patients after diagnosis of the thrombosis, and we did not
influence their decisions. Apart from OT for which redosurgery seems essential, we could not draw any conclusions
about optimal treatment of such early thrombi.
At the beginning of our study, a systematic TEE was
performed at the fifth month (26 patients), but we did not
extend this to the whole population, and so we lack
information on the echographic outcome of all thrombi,
especially in the patients with uneventful course.
Conclusions. This study represents the largest cohort of
patients benefiting from TEE after MMVR and confirms
the high incidence (about 10%) of early postoperative, often
asymptomatic, and NOT.
The course is usually benign for small thrombi (⬍5 mm)
when anticoagulant therapy is adapted. Patients presenting
large NOT develop more complications and require a more
aggressive approach: reintroduction of heparin until oral
anticoagulation is well adjusted, administration of aspirin,
and, in the most threatening large thrombi, redo-surgery.
Reprint requests and correspondence: Dr. Guillaume Laplace,
Service de Cardiologie, Pr Roudaut, Hôpital Cardiologique du
Haut-Lévêque, Avenue Magellan, 33604 Pessac, France. E-mail:
raymond.roudaut@pu.u-bordeaux2.fr.
REFERENCES
1. Malergue MC, Temkine J, Slama M, et al. Intérêts de l’échocardiographie transoesophagienne systématique postopératoire précoce des
remplacements valvulaires mitraux: une étude prospective sur 50
patients. Arch Mal Coeur 1992;85:1299 –304.
2. Dadez E, Iung B, Cormier B, et al. Echocardiographie transoesophagienne précoce après remplacement valvulaire mitral. Signification
des petits échos anormaux. Arch Mal Coeur 1994;87:23–30.
3. Guéret P, Vignon P, Fournier P, et al. Transesophageal echocardiography for the diagnosis and management of nonobstructive thrombosis
of mechanical mitral valve prosthesis. Circulation 1995;91:103–10.
4. Bonnefoy E, Perinetti M, Girard C, et al. Echocardiographie transoesophagienne systématique dans les 24 premières heures postopératoires
après remplacement valvulaire mitral. Arch Mal Coeur 1995;88:315–9.
5. Montorsi P, De Bernardi F, Muratori M, et al. Role of cinefluoroscopy, transthoracic, and transesophageal echocardiography in
patients with suspected prosthetic heart valve thrombosis. Am J
Cardiol 2000;85:58 –64.
6. Iung B, Cormier B, Dadez E, et al. Small abnormal echos after mitral
valve replacement with bileaflet mechanical prothesis predisposing
factors and effect of thromboembolism. J Heart Valve Dis 1993;2:259 –
68.
1290
Laplace et al.
Early Thrombi After Mitral Valve Surgery
7. Bémurat L, Laffort P, Deville C, et al. Management of nonobstructive
thrombosis of prosthetic mitral valve in asymptomatic patients in the
early postoperative period: a study in 20 patients. Echocardiography
1999;16:339 –46.
8. 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. J Am Coll Cardiol 2000;35:739 –46.
9. Edmunds LH, Clark RE, Cohn LH, et al. Guidelines for reporting
morbidity and mortality after cardiac valvular operations. J Thorac
Cardiovasc Surg 1988;96:351–3.
10. Malergue MC, Illouz E, Temkine J, et al. Apport de l’échographie
transoesophagienne mutiplan dans l’étude des prothèses mitrales. Arch
Mal Coeur 1996;89:49 –55.
11. Roudaut R, Labbé T, Marcaggi X, et al. Thrombose de prothèse
valvulaire mécanique. Intérêt de l’échocardiographie transoesophagienne dans l’étude des prothèses mitrales Arch Mal Coeur 1991;84:
503–9.
12. Habib G, Cornen A, Mesana T, et al. Diagnosis of prosthetic heart
valve thrombosis: the respective values of transthoracic and transesophageal Doppler echocardiography. Eur Heart J 1993;14:447–55.
13. Seward JB, Khanderia BK, Freeman WK, et al. Multiplane transesophageal echocardiography: image orientation, examination, technique, anatomic correlation and clinical applications. Mayo Clin Proc
1993;68:523–55.
14. Barbetseas J, Nagueh S, Pitsavos C, et al. Differentiating thrombus
from pannus formation in obstructed mechanical prosthetic valves: an
evaluation of clinical, transthoracic and transesophageal echocardiographic parameters. J Am Coll Cardiol 1998;32:1410 –7.
15. Malergue MC, Abergel E, Bernard Y, et al. Recommandations de la
Société Française de Cardiologie concernant les indications de
l’échocardiographie-Doppler. Arch Mal Coeur 1999;92:1347–79.
16. Özkan M, Kaymaz C, Kirma C, et al. Predictors of left atrial thrombus
and spontaneous echo contrast in rheumatic valve disease before and
after mitral valve replacement. Am J Cardiol 1998;82:1066 –70.
Downloaded From: https://content.onlinejacc.org/ on 09/29/2016
JACC Vol. 43, No. 7, 2004
April 7, 2004:1283–90
17. Virchow R. Phlogose und thrombose in gefessystem. Gessammelte
Abhandlungen zur Wishenschaftlichen Medicin. Frankfurtam-Main,
Germany: Meidiger Sohn and Company, 1856:458 – 63.
18. Alpert JS. The thrombosed prosthetic valve: current recommendations
on evidence from the literature. J Am Coll Cardiol 2003;19:659 –60.
19. Bonow RO, Carabello B, de Leon AC, et al. ACC/AHA 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 Practice Guidelines.
(Committee on Management of Patients With Valvular Heart Disease). Circulation 1998;98:1949 –84.
20. Roudaut R, Lafitte S, Lorient-Roudaut MF, et al. Fibrinolysis of
mechanical prosthetic valve thrombosis: a single center study of 127
cases. J Am Coll Cardiol 2003;41:653–8.
21. Özkan M, Kaymaz C, Kirma C, et al. Intravenous thrombolytic
treatment of mechanical prosthetic valve thrombosis: a study using
serial transesophageal echocardiography. J Am Coll Cardiol 2000;35:
1881–9.
22. Shapira Y, Herz I, Vaturi M, et al. Thrombolysis is an effective and
safe therapy in stuck bileaflet mitral valves in the absence of high-risk
thrombi. J Am Coll Cardiol 2000;35:1874 –80.
23. Lengyel M, Fuster V, Keltai M, et al. Guidelines for management of
left-sided prosthetic valve thrombosis: a role for thrombolytic therapy.
J Am Coll Cardiol 1997;30:1521–6.
24. Silber H, Khan S, Matloff J, et al. The St. Jude valve: thrombolysis as
the first line of therapy for cardiac valve thrombosis. Circulation
1993;87:30 –7.
25. Temkine J, Malergue MC, Benabho T, Lecompte Y. Thombose
postopératoire asymptomatique d’une valve de Saint Jude mitrale:
traitement par fibrinolyse avec rtPA. Arch Mal Coeur 1991;84:
413–7.
26. Lengyel M, Vegh G, Vandor L. Thrombolysis is superior to heparin
for non-obstructive mitral mechanical valve thrombosis. J Heart Valve
Dis 1999;8:167–73.