Content uploaded by Zivojin S Jonjev
Author content
All content in this area was uploaded by Zivojin S Jonjev on Dec 22, 2018
Content may be subject to copyright.
Mitral Valve Replacement Is a Viable Alternative
to Mitral Valve Repair for Ischemic Mitral
Regurgitation: A Case-Matched Study
Vincent Chan, MD, MPH, Marc Ruel, MD, MPH, and Thierry G. Mesana, MD, PhD
Division of Cardiac Surgery and Department of Epidemiology and Community Medicine, University of Ottawa, Ottawa, Ontario,
Canada
Background. Comparisons of mitral valve repair with
mitral valve replacement for ischemic mitral regurgita-
tion (IMR) have been limited by differences in preoper-
ative and operative characteristics of patients undergoing
these two types of surgical treatment. We performed a
propensity-based, case-matched analysis to examine
whether patients who undergo mitral valve repair and
those who undergo mitral valve replacement for IMR
have similar long-term outcomes.
Methods. We compared 65 patients who underwent
mitral valve replacement for IMR between 2001 and 2010
with 65 patients who underwent mitral repair during the
same period on the basis of age, concomitant coronary
bypass grafting, gender, left ventricular function, preop-
erative pulmonary hypertension, and urgency of opera-
tion. Mitral replacement involved preservation of the
subvalvular apparatus. The mean study follow-up period
was 2.5 ⴞ2.1 years.
Results. Two patients who underwent mitral valve
repair died at 30 days postoperatively and three patients
died after valve replacement. Late survival was the same
in the two groups (pⴝ0.4). Recurrent mitral regurgita-
tion (MR) (grade 2ⴙor higher) at late follow-up was
observed in 15 patients (23%) after repair; however, only
1 patient (2%) had MR with a grade of more than 2ⴙ.
Mitral valve repair was more commonly associated
with recurrent MR (grade 2ⴙor higher) than was
mitral valve replacement (pⴝ0.04). Patients in both
groups had similar freedom from valve-related compli-
cations and similar left ventricular function at fol-
low-up (both p>0.2).
Conclusions. Mitral valve replacement remains a viable
option for the treatment of IMR. Although mitral valve
repair effectively protects against persistent or recurrent
moderate-to-severe MR, mitral valve replacement pro-
vides better freedom from mild-to-moderate MR in this
population, with a low incidence of valve-related com-
plications. Notably, there was no significant difference in
left ventricular function between the valve-repair and
valve-replacement groups at follow-up.
(Ann Thorac Surg 2011;92:1358–66)
© 2011 by The Society of Thoracic Surgeons
Ischemic mitral regurgitation (IMR) is common and
associated with poor outcomes. It is observed in up to
50% of patients presenting with acute coronary syn-
dromes and in 19% of patients undergoing elective car-
diac catheterization [1]. Approximately half of patients
with IMR and recent acute coronary syndromes will
develop associated heart failure [2]. Overall, patients
with IMR have much worse long-term survival and
functional status than do patients with coronary disease
without IMR [3, 4].
The management of IMR is complex and depends in
part on the severity of preoperative mitral regurgitation
(MR). In patients with moderate MR (grade 2⫹), there is
controversy about whether mitral valve surgery concom-
itant with coronary artery bypass grafting (CABG) is
better than CABG alone [5–8]. Although some advocate
CABG alone [5, 6], others have found that CABG with
mitral valve surgery may reduce postoperative MR and
cardiac-related mortality [7–9]. It is generally agreed that
MR that is at least moderate to severe (grades 3⫹or 4⫹)
requires mitral valve intervention, but whether mitral
valve repair or mitral valve replacement is the better type
of surgical procedure remains a subject of debate [10].
In some series, mitral valve repair was found to pro-
vide better survival and to cause fewer long-term com-
plications than mitral valve replacement. Grossi and
colleagues found that at 5 years after surgery, the rate of
complication-free survival of patients who underwent
mitral repair for IMR was 64%, as compared with 47% for
patients who underwent mitral replacement [11]. A meta-
analysis of studies published between 1999 and 2009 also
indicated that mitral repair conferred better short- and
long-term survival than did mitral replacement [12]. The
Accepted for publication May 16, 2011.
Presented at the Forty-seventh Annual Meeting of The Society of Thoracic
Surgeons, San Diego, CA, Jan 31–Feb 2, 2011.
Address correspondence to Dr Ruel, University of Ottawa Heart Institute,
3403-40 Ruskin St, Ottawa, ON, Canada, K1Y 4W7; e-mail: mruel@
ottawaheart.ca.
Dr Ruel discloses that he has financial relationships
with Sanofi-Aventis, Bristol-Myers Squibb, Edwards
Lifesciences, and Medtronic, Inc.
© 2011 by The Society of Thoracic Surgeons 0003-4975/$36.00
Published by Elsevier Inc doi:10.1016/j.athoracsur.2011.05.056
ADULT CARDIAC
main benefit of mitral repair is thought to be related to
the improvement in left-ventricular (LV) function that
occurs with preservation of the subvalvular apparatus,
which historically was often removed at the time of mitral
valve replacement [13].
However, the long-term benefit of mitral repair may
not apply to all patients with IMR. In several series, no
short- or long-term survival benefit of mitral repair over
mitral replacement was identified [14–17]. Recurrent MR
is an important problem following repair, and may be
found in approximately 30% of patients at 6 months
postoperatively [18]. Moreover, contemporary mitral
valve replacement is done with better bioprostheses than
were used in the past, as well as with preservation of the
subvalvular apparatus, which may reduce the LV dys-
function that was historically associated with mitral
replacement [19, 20].
The question of whether IMR should be corrected with
mitral valve repair or replacement is the basis for a
randomized study sponsored by the National, Heart,
Lung and Blood Institute [21]. That study is currently still
enrolling patients, and the clinical dilemma related to
this important question is ongoing. At present, the data
guiding clinicians come from published studies that are
limited by differences in patients who receive valve
repair versus valve replacement [15–17]. Additionally,
previous attempts at matched comparisons of mitral
repair and mitral replacement were significantly limited
by the lack of long-term assessment of recurrent MR and,
importantly, of LV function [22]. We therefore conducted
a propensity-based, case-matched follow-up analysis of
130 patients who underwent either mitral repair (n ⫽65)
or mitral replacement (n ⫽65) for IMR. These patients
were matched on the basis of age, concomitant CABG,
gender, LV function, preoperative pulmonary hyperten-
sion, and urgency of operation. We compared mitral
valve repair with mitral valve replacement for the treat-
ment of IMR with regard to survival, freedom from
recurrent MR (grade 2⫹or higher and grade 3⫹or
higher), and postoperative LV function.
Patients and Methods
Ethics Approval and Funding
The study was approved by the institutional review
board of The University of Ottawa Heart Institute. The
University of Ottawa Heart Institute Heart Valve Clinic
receives unrestricted research funding from Edwards
Lifesciences, Medtronic, and On-X Life Technologies.
Patient Population
Between 2001 and 2010, 230 patients underwent mitral
valve surgery at our institution for the correction of
severe (grade 3⫹or 4⫹), symptomatic IMR. Ischemic MR
was identified by preoperative echocardiography as well
as intraoperative assessment by the attending surgeon.
Of these 230 patients, 69 patients underwent mitral valve
replacement and 161 patients underwent mitral valve
repair. Sixty-five patients who presented for surgery with
the intent to undergo mitral valve replacement were
matched on the basis of age, concomitant CABG, gender,
LV function, preoperative pulmonary hypertension, and
urgency of operation with 65 patients who underwent
mitral valve repair. The two groups of patients constitute
the basis for the present study.
Operative Characteristics
All 130 patients in the study underwent mitral valve
surgery via a median sternotomy, with the use of cardio-
pulmonary bypass (CPB), mild systemic hypothermia,
and cold-blood cardioplegia. Downsizing ring annulo-
plasty was used for all patients who underwent mitral
valve repair. Additional techniques for mitral valve re-
pair included edge-to-edge repair in 34 patients (52%),
chordal transfer in 3 patients (5%), and the use of poly-
tetrafluroethylene chords in 2 patients (3%). Annulo-
plasty bands or rings used in the study included the
Medtronic Futureband (Medtronic, Minneapolis, MN) in
49 patients (75%), the Duran Ancore (Medtronic) in 7
patients (10%), the Edwards Physio Ring (Baxter Health-
care Corp, Irvine, CA) in 5 patients (8%), and the Cos-
grove-Edwards Annuloplasty System (Baxter) in 4 pa-
tients (6%). The mean annuloplasty ring/band size was
28.4 ⫾1.6 mm (median, 28 mm; range, 26 to 32 mm).
Mitral valve replacement was performed with preserva-
tion of the subvalvular apparatus. The posterior mitral
valve leaflet was left intact in all patients undergoing
mitral replacement. In 27 of the patients undergoing
mitral replacement, the anterior leaflet of the valve was
partly or completely detached from the mitral annulus
and divided in the middle at the 12 o’clock position, and
the leftward portion of the anterior leaftlet was plicated
to the anterolateral commissure with a pledgetted 4-0
polypropylene suture. The rightward portion of the an-
terior mitral leaflet was similarly plicated to the postero-
medial commissure with a pledgetted 4-0 polypropylene
suture. In the other 38 patients undergoing mitral valve
replacement, the middle portion of the anterior leaflet
was resected and the remaining leaflet tissue was pli-
cated with the individual valve sutures. Prostheses im-
planted as mitral valve replacements included the
Medtronic Hancock II, (Medtronic) in 46 patients (71%),
On-X valve (On-X Life Technologies, Austin, TX) in 17
patients (26%), and Edwards Porcine (Baxter) in 2 pa-
tients (3%). The mean mitral prosthesis size was 27.9 ⫾
2.5 mm (median, 27 mm). Concomitant CABG was per-
formed on main coronary vessels or branches of 1.5 mm
or more in diameter that displayed luminal stenosis of
70% or more on preoperative angiography.
Assessment of Preoperative and Intraoperative Mitral
Regurgitation
Preoperatively, all patients underwent M-mode, two-
dimensional, and Doppler transthoracic or transesopha-
geal echocardiography or both for measurement of their
cardiac dimensions, transvalvular pressure gradients,
and severity of valvular regurgitation, as recommended
by the American Society of Echocardiography (ASE)
[23–25]. The mechanism of MR was classified according
1359
Ann Thorac Surg CHAN ET AL
2011;92:1358–66 MITRAL REPAIR VERSUS REPLACEMENT FOR IMR
ADULT CARDIAC
to the system described by Carpentier [26], using both
preoperative echocardiographic data and information
obtained during intraoperative assessment of each pa-
tient’s mitral valve by the attending surgeon. The etiol-
ogy of MR was considered to be of type I when there was
at least mild annular dilation and of type IIIb when there
was at least mild posterior leaflet restriction. In patients
who underwent mitral valve repair, MR exceeding grade
1⫹or a mean mitral intravalvular pressure gradient
exceeding 5 mm Hg determined intraoperatively were
considered indications for further mitral valve interven-
tion. One patient had persistent MR of grade 3⫹after
initial mitral valve repair with an isolated ring annulo-
plasty. The valve was successfully re-repaired with the
addition of edge-to-edge sutures after a return to CPB.
Follow-Up
Patients were assessed annually in a dedicated valve
clinic. The procedures conducted at clinic visits consisted
of a history and physical examination, electrocardiogram,
chest radiograph, complete blood count, serum chemis-
tries, and determination of the international normalized
ratio (INR) (when applicable). Prosthesis-related compli-
cations were recorded according to Akins and colleagues’
guidelines for reporting morbidity and mortality after
cardiac valve interventions [27]. The mean postoperative
follow-up period was 2.5 ⫾2.1 years (median, 2.1 years).
Postoperative echocardiographic follow-up was also
done, at 1, 3, 6, and 12 months, on all patients who had
mitral valve repair. Beyond 12 months postroperatively,
echocardiograms where made every 2 to 3 years or when
clinically indicated. The mean echocardiographic fol-
low-up for patients who underwent mitral valve repair
was 2.1 ⫾2.0 years (median, 1.5 years).
Statistical Analyses
Data were imported and analyzed with the Stata 10.1
statistical software program (Stata, College Station, TX).
Patients who underwent mitral valve repair were com-
pared with those who underwent mitral valve replace-
ment with respect to demographic, echocardiographic,
and operative parameters. As noted earlier, the 65 pa-
tients in the study who underwent mitral valve replace-
ment were matched with 65 patients who underwent
mitral repair on the basis of age, concomitant CABG,
gender, preoperative LV function, preoperative pulmo-
nary hypertension, and urgency of operation. These
patients were matched through the calculation of a global
propensity score, which was calculated on the basis of the
above listed variables.
Continuous data are described as mean ⫾standard
deviation. A
2
or Fisher’s exact test was used for cate-
gorical variables, whereas a one-way analysis of variance
was used for continuous variables. A log-rank test was
used to determine equality of risk factors for recurrent
MR of grade 2⫹or higher. The proportional hazards
assumption was verified with Schoenfeld residuals.
With 130 patients, the power to detect a hazard ratio of
1.5 with a standard deviation of 0.5 via a two-tailed test
was 68%. The estimated effect size was selected on the
basis of both the study data and previously published
data [15–17]. Preoperative and postoperative LV function
were divided into four grades based on increasing LV
dysfunction: grade I, LV ejection fraction (LVEF) exceed-
ing 50%; grade II, LVEF of 35% to 50%; grade III, LVEF of
20% to 34%; and grade IV, LVEF below 20%. Left ventric-
ular function was compared among the groups by deter-
mining the change in LV grade following surgery. Im-
provement in LV function was defined as a pre- to
postoperative decrease in LV grade of 1 or more. Deteri-
oration in LV function was defined as a pre- to postop-
erative increase in LV grade of 1 or more. A logistic
regression model was used to determine risk factors
associated with a postoperative decrease in LV grade.
Statistical significance was set at p⬍0.05.
Results
Characteristics of Patients Who Underwent Mitral
Valve Repair and Those Who Underwent Mitral Valve
Replacement for IMR
The demographic and echocardiographic characteristics
of patients who underwent mitral valve repair or replace-
ment because of IMR are shown in Table 1. The patients’
mean age at operation was 67.7 ⫾9.1 years. Patients in
both the mitral repair and mitral replacement groups
were similar with regard to age, concomitant CABG,
gender, and preoperative pulmonary hypertension. The
patients in the two groups were also similar with regard
to echocardiographic parameters such as preoperative
left atrial (LA) size and both LV end-diastolic diameter
(LVEDD) and LV end-systolic diameter (LVESD).
Concomitant CABG was performed in 49 patients
(75%) who underwent mitral valve repair and 56 patients
(86%) who underwent mitral valve replacement. The
reasons for nonperformance of concomitant CABG at the
time of mitral valve surgery were percutaneous coronary
intervention in 11 patients (7 who had mitral repair and 4
who had mitral replacement), previous CABG with pat-
ent bypass grafts in 8 patients (4 who had repair and 4
who had replacement), and small coronary target or
distal lesions in 6 patients (5 who had repair and 1 who
had replacement).
Posterior mitral valve leaflet restriction (type IIIb Car-
pentier class of MR [26]), was the most common etiology
of MR (Table 1). Among the 13 patients with type II valve
prolapse who underwent mitral valve replacement, 11
had papillary muscle rupture. The remaining 2 patients
with type II valve prolapse who underwent valve replace-
ment, and 3 patients with type II valve prolapse who
underwent valve repair, had elongated or torn chordae
tendineae. Mitral valve repair in the latter 3 patients
involved a combination of chordal transfer, the use of
polytetrafluroethylene chords, and downsizing ring an-
nuloplasty. Overall, the matching propensity scores in
the mitral replacement and mitral repair groups were
similar.
1360 CHAN ET AL Ann Thorac Surg
MITRAL REPAIR VERSUS REPLACEMENT FOR IMR 2011;92:1358– 66
ADULT CARDIAC
Long-Term Survival of Patients With IMR
Thirty-day mortality across both groups of patients in the
study was 4% (5 of 130 patients), and was not significantly
different in the two groups (Table 1). Twenty-two late
deaths were observed during follow-up, of which 10 oc-
curred after mitral valve repair and 12 after mitral valve
replacement. Causes of death included cancer (n ⫽5),
cardiac causes (n ⫽5), gastrointestinal disorders (n ⫽3),
multiorgan failure (n ⫽2), respiratory failure (n ⫽2),
ruptured abdominal aortic aneurysm (n ⫽1), other noncar-
diac causes (n ⫽2), and unknown causes (n ⫽2). Long-term
survival was similar in the two study groups, with a 5-year
survival of 79.3 ⫾6.1% for patients who underwent mitral
repair and of 60.6 ⫾8.8% for those who underwent mitral
replacement, (p⫽0.4) (Fig 1). Among patients with type I
annular dilation or type IIIb leaflet restriction (ie, excluding
patients with a type II etiology of MR), 5-year survival was
also not different in the mitral repair and mitral replace-
ment groups (79.3 ⫾6.1% vs 68.7 ⫾10.1%, respectively, p⫽
0.7). Five-year freedom from known cardiac-related mortal-
ity was 96.2 ⫾2.6% and 95.1 ⫾2.8% for patients who
underwent mitral repair and mitral replacement, respec-
tively (p ⫽0.6).
Likelihood of Recurrence of MR of Grade 2⫹or
Higher and Grade 3⫹or Higher in Patients
Undergoing Mitral Repair or Mitral Replacement
In follow-up, 16 patients, of whom 15 had undergone mitral
valve repair and 1 had undergone mitral valve replacement,
developed recurrent MR of grade 2⫹or higher. Only 1 of
the 15 patients who had undergone mitral repair had
developed recurrent MR of grade 3⫹or 4⫹at the time of
follow-up. The 1 patient who developed recurrent severe
MR of grade 4⫹after mitral replacement had endocarditis
of the prosthetic valve and underwent reoperation. Of the
14 patients who developed recurrent grade 2⫹MR after
mitral valve repair, 10 had grade 4⫹MR preoperatively and
4 others had grade 3⫹MR preoperatively. Six-month free-
dom from recurrent MR of grade 2⫹or higher was 95.0 ⫾
3.5% after mitral valve repair and 100% following mitral
valve replacement, whereas 5-year freedom from recurrent
Table 1. Patient Characteristics
Variable
Mitral Valve
Repair
(N ⫽65)
Mitral Valve
Replacement
(N ⫽65)
p-
Value
Demographics
Age (years) 66.9 ⫾8.6 68.5 ⫾9.5 0.3
Female gender 20 (31%) 22 (34%) 0.7
Concomitant CABG
a
0 16 (25%) 10 (15%)
1 22 (34%) 13 (20%)
2 15 (23%) 16 (25%)
ⱖ3 12 (18%) 26 (40%) 0.1
Echocardiographic
Variables
Preoperative LV
ejection fraction
(%)
37.4 ⫾9.3 37.0 ⫾12.9 0.9
Preoperative LV
grade
b
I 9 (14%) 18 (28%)
II 20 (31%) 23 (35%)
III 25 (38%) 15 (23%)
IV 11 (17%) 9 (14%) 0.2
Pulmonary
hypertension
c
33 (51%) 36 (55%) 0.7
Preoperative LA Size
(mm)
47.5 ⫾6.5 47.3 ⫾8.5 0.9
Preoperative LVEDD
(mm)
60.9 ⫾7.7 58.4 ⫾9.2 0.2
Preoperative LVESD
(mm)
47.0 ⫾10.1 42.3 ⫾13.1 0.1
Classification of MR
d
Type I 25 (37%) 15 (23%) 0.06
Type II 3 (5%)
e
13 (20%)
f
0.008
Type IIIb 48 (74%) 48 (74%) 0.9
Operative Variables
Elective surgery 44 (68%) 30 (46%) 0.02
Thirty-day mortality 2 (3%) 3 (5%) 0.7
Global propensity
score
0.53 ⫾0.09 0.50 ⫾0.08 0.1
a
Test of proportions using a dichotomous variable.
b
LV Grade I ⫽
ejection fraction ⬎50%; Grade II ⫽ejection fraction 35–50%; Grade III ⫽
ejection fraction 20–34%; Grade IV ⫽ejection fraction ⬍20%.
c
De-
fined as preoperative right ventricular systolic pressure greater than 45
mm Hg.
d
Defined according to the classification of MR described by
Carpentier [26].
e
The 3 patients with type II valve prolapse who
underwent mitral valve repair had elongated or torn chor-
dae.
f
Among the 13 patients with type II valve prolapse who under-
went mitral valve replacement, 11 had papillary muscle rupture whereas
2 had elongated or torn chordae.
CABG ⫽coronary artery bypass grafting; LA ⫽left atrium; LV ⫽
left ventricle; LVEDD ⫽left ventricle end-diastolic diameter;
LVESD ⫽left ventricle end-systolic diameter; MR ⫽mitral
regurgitation.
Fig 1. Long-term survival after mitral repair or replacement. This
figure represents all-cause mortality for mitral valve surgery for
ischemic mitral regurgitation (I MR). In total, 22 deaths were ob-
served in this cohort: 10 after mitral valve repair and 12 after mitral
valve replacement. Causes of death included cancer (n ⫽5), cardiac
causes (n ⫽5), gastrointestinal disorders (n ⫽3), multiorgan failure
(n ⫽2), respiratory failure (n ⫽2), ruptured abdominal aortic an-
eurysm (n ⫽1), other noncardiac causes (n ⫽2), and unknown
causes (n ⫽2). Concomitant coronary artery bypass grafting
(CABG) was performed in 46 patients who underwent mitral valve
repair and 55 patients who underwent mitral valve replacement.
1361Ann Thorac Surg CHAN ET AL
2011;92:1358–66 MITRAL REPAIR VERSUS REPLACEMENT FOR IMR
ADULT CARDIAC
MR of grade 2⫹or higher was 41.4 ⫾14.8% following mitral
valve repair and 85.7 ⫾13.2% after mitral valve replacement
(p⫽0.04) (Fig 2). Five-year freedom from recurrent MR of
grade 3⫹or higher was 90.9 ⫾8.7% after mitral repair and
85.7 ⫾13.2% after mitral replacement (p⫽0.9). The number
of grafts performed at mitral valve surgery was inversely
associated with the recurrence of MR of grade 2⫹or higher
(hazard ratio [HR] 0.5 per additional graft, 95% confidence
interval [CI] 0.3 to 0.9, p⫽0.02) (Table 2).
Lack of Association of Etiology of MR and of Size of
Annuloplasty Band or Ring With Development of
Recurrent MR of Grade 2⫹or Higher in Patients
Undergoing Mitral Valve Repair for IMR
Patients who developed recurrent MR of grade 2⫹or
higher after mitral valve repair were similar in terms of age,
gender, and preoperative LV function to those who under-
went repair without developing MR (Table 3). The etiology
of IMR, whether type I or type IIIb according to the
Carpentier classification, was not associated with recurrent
MR of grade 2⫹or higher (p⬎0.6 for both types of
etiology). Nor was the ring or band size used in annulo-
plasty associated with recurrent MR of grade 2⫹or higher
(p⫽0.9).
Valve-Related Complications and Late LV function
With Mitral Valve Repair and Mitral Valve
Replacement
Two mitral valve-related complications occurred in patients
in the study who underwent valve replacement. One was a
Fig 2. Freedom from recurrent mitral regurgi-
tation (MR) after mitral repair or replacement.
Recurrent MR was determined through post-
operative echocardiographic assessments. All
patients underwent transthoracic echocardiog-
raphy following repair according to clinic
guidelines. In patients with multiple echocar-
diographic assessments, the earliest date in
which recurrent MR developed was used for
analysis. However, if recurrent MR was not
observed, the most recent date of echocardiog-
raphy was utilized. Figure 2A describes recur-
rent MR (ⱖ2⫹) whereas Figure 2B describes
recurrent MR 3⫹or 4⫹.
Table 2. Risk Factors Associated With Recurrent MR (ⱖ2⫹)
Risk Factor Hazard Ratio 95% CI p-Value
Significant Covariates
Number of coronary grafts
(per additional graft)
0.5 0.3–0.9 0.02
Mitral valve repair 8.0 1.1–62.4 0.04
Non-Significant Covariates
Age (per increasing year) 1.0 0.9–1.0 0.5
Female gender 0.6 0.1–2.5 0.4
Preoperative LV function
(per increasing grade)
a
1.2 0.6–2.4 0.6
a
LV Grade I ⫽ejection fraction ⬎50%; Grade II ⫽ejection fraction
35–50%; Grade III ⫽ejection fraction 20–34%; Grade IV ⫽ejection
fraction ⬍20%.
CI ⫽confidence interval; LV ⫽left ventricle; MR ⫽mitral
regurgitation.
1362 CHAN ET AL Ann Thorac Surg
MITRAL REPAIR VERSUS REPLACEMENT FOR IMR 2011;92:1358– 66
ADULT CARDIAC
case of endocarditis (mentioned above) in a prosthetic
mechanical mitral valve, requiring reoperation at 24
months after initial mitral valve replacement surgery. The
other was a case of structural deterioration of a porcine
valve, which presented as severe mitral stenosis that re-
quired reoperation with the insertion of a mechanical
prosthesis at 47.5 months after initial valve replacement. No
hemorrhage or thromboembolism was observed during the
study period in patients who underwent valve replacement.
One case of peripheral embolism occurred at 73.2 months
after mitral valve repair in a patient who was being treated
with warfarin for chronic atrial fibrillation, and another
patient in the mitral repair group developed MR of grade
4⫹at 5.5 months after valve surgery, as a result of bacterial
endocarditis. This patient subsequently underwent suc-
cessful mitral valve replacement.
Seven-year freedom from valve-related complications
was 80.0 ⫾17.9% after mitral valve repair and 84.3 ⫾9.2%
after mitral valve replacement (p⫽0.2). The proportion
of patients whose LV function improved or deteriorated
postoperatively as compared with its preoperative level
did not differ with either mitral valve repair or replace-
ment (both p⬎0.5) (Table 4). Nor was mitral valve repair
associated with a decrease in postoperative LV grade
(odds ratio [OR] 1.0, 95% CI 0.2 to 3.8, p⫽0.9), although
concomitant CABG was associated with such a decrease
(OR 0.2, 95% CI 0.04 to 0.8, p⫽0.02) (Table 5).
Comment
In this study we compared 65 patients who underwent
mitral valve repair with 65 patients who underwent mitral
valve replacement for the treatment of IMR. These patients
were matched according to age, concomitant CABG, gen-
der, preoperative LV function, preoperative pulmonary
hypertension, and urgency of operation. The study was
done to compare mitral valve repair with mitral valve replace-
ment in terms of long-term postoperative survival, freedom
from recurrent MR, and postoperative LV function.
The short-term and long-term mortalities of patients in
the study compare with those in previously published
series. The published operative mortality of patients
undergoing CABG with concomitant mitral valve repair
or replacement for IMR ranges from 4% to 14% [7, 14–16,
28], and their reported unadjusted 5-year survival ranges
from 44% to 64% [29–30]. Although some investigators
have found that mitral valve repair is associated with
better survival than is mitral valve replacement [31–32],
others have found no difference [14–16]. Gillinov and
associates surmised that in patients with very severe LV
dysfunction, the surgical approach to the mitral valve is
less important given the poor overall survival in this
population [22]. In the present study we observed no
difference in overall survival in the mitral valve repair and
mitral valve replacement groups. We also found that the
freedom from cardiac-related death after either valve repair
or replacement was greater than 90%. Lower values for
Table 3. Risk Factors for the Development of Recurrent MR
(ⱖ2⫹) in Mitral Valve Repair Patients
Risk Factor Hazard Ratio 95% CI p-Value
Age (per increasing years) 0.99 0.92–1.06 0.7
Concomitant CABG
a
0.3 0.1–1.1 0.07
Female gender 0.5 0.1–2.4 0.4
Preoperative LV grade
b
1.1 0.5–2.3 0.7
Classification of MR
c
Type I 1.0 0.3–3.3 0.9
Type IIIb 1.4 0.4–5.4 0.6
Annuloplasty size (per
increasing mm)
1.0 0.7–1.5 0.9
a
Entered into model as an ordinal variable.
b
LV Grade I ⫽ejection
fraction ⬎50%; Grade II ⫽ejection fraction 35–50%; Grade III ⫽ejection
fraction 20–34%; Grade IV ⫽ejection fraction ⬍20%.
c
Defined ac-
cording to the classification of MR described by Carpentier [26].
CABG ⫽coronary artery bypass grafting; CI ⫽confidence inter-
val; LV ⫽left ventricle; MR ⫽mitral regurgitation.
Table 4. Change in LV Function AfterMitral Valve Repair or
Replacement
a
Type of Surgery
LV
Improvement
(N ⫽37)
No Change
in LV
Function
(N ⫽46)
LV
Deterioration
(N ⫽47)
Mitral valve repair 16 23 26
Mitral valve
replacement
21 23 21
pValue 0.5 0.9 0.6
a
Improvement in LV function was defined as an increase of ⱖ1 grade
postoperatively compared to preoperatively. Similarly, deterioration of
LV function was defined as a decrease of ⱖ1 grade postoperatively
compared to preoperatively.
LV Grade I ⫽ejection fraction ⬎50%; Grade II ⫽ejection fraction
35–50%; Grade III ⫽ejection fraction 20–34%; Grade IV ⫽ejection
fraction ⬍20%.
LV ⫽left ventricle.
Table 5. Risk Factors Associated With a Decrease in
Postoperative LV Function After Mitral Valve Repair or
Replacement
Risk Factor Odds Ratio 95% CI p-value
Significant covariates
Concomitant CABG
a
0.2 0.04–0.8 0.02
Non-Significant covariates
Age (per increasing year) 1.02 0.95–1.10 0.5
Female gender 0.5 0.1–1.9 0.3
Preoperative LV function
(per increasing grade)
b
1.4 0.7–2.9 0.3
Mitral valve repair 1.0 0.2–3.8 0.9
a
Entered into logistic regression model as a dichotomous variable.
b
Im-
provement in LV function was defined as a decrease of ⱖ1 grade
postoperatively compared to preoperatively. Similarly, worsening of LV
function was defined as an increase of ⱖ1 grade postoperatively com-
pared to preoperatively.
LV Grade I ⫽ejection fraction ⬎50%; Grade II ⫽ejection fraction
35–50%; Grade III ⫽ejection fraction 20–34%; Grade IV ⫽ejection
fraction ⬍20%.
CABG ⫽coronary artery bypass grafting; CI ⫽condidence inter-
val; LV ⫽left ventricle.
1363Ann Thorac Surg CHAN ET AL
2011;92:1358–66 MITRAL REPAIR VERSUS REPLACEMENT FOR IMR
ADULT CARDIAC
freedom from cardiac-related death have been reported in
other series [7–9]. This difference may reflect the relatively
high proportion of noncardiac causes of death in our study.
Four percent of our patients died of cancer and 6% died of
pulmonary, gastrointestinal, or vascular causes.
In this study, mitral valve repair was associated with
higher rates of recurrent MR of grade 2⫹or more than
was mitral valve replacement. Six-month freedom from
recurrent MR of grade 2⫹or higher was 95.0 ⫾3.5% after
repair, which compares favorably with that in other
series [18]. Although only one patient in our study
developed MR exceeding grade 2⫹after mitral valve
repair, the 5-year freedom from recurrent MR of grade
2⫹or higher was 41%. The recurrence of MR may be
related to incomplete LV reverse remodeling [31]. Preop-
erative LVEDD enlargement of more than 65 to 70 mm is
associated with poor LV remodeling and recurrent MR
[32, 33]. In patients who developed recurrent MR of grade
2⫹or more after repair, we did not find that preoperative
LV size was a significant risk factor. However, the num-
ber of coronary grafts done in our study was an impor-
tant determinant of recurrent MR of grade 2⫹or higher.
These observations suggest that the degree of potential
LV recovery as a result of revascularization is a more
important predictor of recurrent MR of grade 2⫹or
higher than is absolutely measured preoperative LV size.
Not surprisingly, a small annuloplasty band or ring size
alone failed to protect against recurrent MR, which is in
keeping with observations made by others [18].
Another important observation in our study was that
mitral valve replacement was associated with few valve-
related complications. Despite the single case of pros-
thetic-valve endocarditis and single case of structural
deterioration of a porcine bioprosthesis, the overall rate
of occurrence of endocarditis and structural valve dete-
rioration in the study was lower than that in other
published series [34, 35]. Only 26% of the patients in the
study received a mechanical prosthesis, which may
partly explain the lack of hemorrhagic or thromboem-
bolic events. Low rates of thromboembolism have been
reported with the On-X valve, which was the only me-
chanical valve used in the study [36]. The case of periph-
eral embolism in the mitral valve repair group occurred
in a patient with chronic atrial fibrillation, which is a
well-established risk factor for thromboembolism [37].
Study Limitations
Patients who underwent mitral valve repair were matched
with those who underwent mitral valve replacement on the
basis of a calculated propensity score. The two groups may
have differed with regard to important risk factors not
included in the propensity score. Matching also fails to
completely eliminate important but statistically nonsignifi-
cant differences between groups. Although there was no
difference between the two groups in the study with regard
to concomitant CABG, there was a trend toward more
coronary grafts in patients who underwent mitral valve
replacement than in those who had mitral valve repair.
Because the study compared mitral valve repair with mitral
valve replacement in patients with severe preoperative MR,
no comparisons were made involving patients with moder-
ate MR who underwent CABG without mitral valve sur-
gery. Also, because noninvasive measurements of myocar-
dial viability were not available preoperatively, the benefit
of CABG in patients with LV dysfunction remains undeter-
mined. This last point is not only relevant for postoperative
LV function, but also for survival and the development of
recurrent MR. Few valve-related complications were ob-
served in the study, but more complications would have
been observed with longer follow-up, larger numbers of
patients, or both. The power to detect a difference in overall
survival or cardiac-related survival was also limited and can
similarly be increased with longer follow-up.
Conclusions
Mitral valve replacement remains a viable option for treat-
ing IMR. Although mitral valve repair effectively protects
against persistent or recurrent moderate to severe MR,
mitral replacement provides better freedom from mild to
moderate MR in the population with IMR, with a low
incidence of valve-related complications. Notably, the mi-
tral replacement and mitral repair groups showed no sig-
nificant difference in LV function at follow-up.
References
1. Hickey MS, Smith LR, Muhlbaier LH, et al. Current progno-
sis of ischemic mitral regurgitation: implications for future
management. Circulation 1988;78:151–9.
2. Trichon BH, Felker GM, Shaw LK, et al. Relation of fre-
quency and severity of mitral regurgitation to survival
among patients with left ventricular systolic dysfunction and
heart failure. Am J Cardiol 2003;91:538 – 43.
3. Bursi F, Enriquez-Sarano M, Nkomo VT, et al. Heart failure and
death after myocardial infarction in the community: the emerging
role of mitral regurgitation. Circulation 2005;111: 295–301.
4. Lamas GA, Mitchell GF, Flaker GC, et al. Clinical signifi-
cance of mitral regurgitation after acute myocardial infarc-
tion. Circulation 1997;96:827–33.
5. Ryden T, Bech-Hanssen O, Brandrup-Wognsen G, Nilsson
F, Svensson S, Jeppsson A. The importance of grade 2
ischemic mitral regurgitation in coronary artery bypass
grafting. Eur J Cardiothorac Surg 2001;20:276 – 81.
6. Duarte IG, Shen Y, MacDonald MJ, Jones EL, Craver JM,
Guyton RA. Treatment of moderate mitral regurgitation and
coronary disease by coronary bypass alone: late results. Ann
Thorac Surg 1999;68:426 –30.
7. Aklog L, Filsoufi F, Flores KQ, et al. Does coronary artery
bypass grafting alone correct moderate ischemic mitral re-
gurgitation? Circulation 2001;104:168 –75.
8. Lam BK, Gillinov AM, Blackstone EH, et al. Importance of
moderate ischemic mitral regurgitation. Ann Thorac Surg
2005;79:462–70.
9. Fattouch K, Sampognaro R, Speziale G, et al. Impact of
moderate ischemic mitral regurgitation after isolated coronary
artery bypass grafting. Ann Thorac Surg 2010;90:1187–94.
10. Bonow RO, Carabello BA, Chatterjee K, et al. 2008 focused
update incorporated into the ACC/AHA 2006 guidelines for the
management of patients with valvular heart disease: a report of
the American College of Cardiology/American Heart Associa-
tion Task Force on Practice Guidelines (Writing Committee to
revise the 1998 guidelines for the management of patients with
valvular heart disease). J Am Coll Cardiol 2008;52:676 – 85.
11. Grossi EA, Goldberg JD, LaPietra A, et al. Ischemic mitral
valve reconstruction and replacement: comparison of long-
term survival and complications. J Thorac Cardiovasc Surg
2001;122:1107–24.
1364 CHAN ET AL Ann Thorac Surg
MITRAL REPAIR VERSUS REPLACEMENT FOR IMR 2011;92:1358– 66
ADULT CARDIAC
12. Vassileva CM, Boley T, Markwell S, Hazelrigg S. Meta-
analysis of short-term and long-term survival following
repair versus replacement for ischemic mitral regurgitation.
Eur J Cardiothorac Surg 2011;39:295–303.
13. David TE, Burns RJ, Bacchus CM, Druck MN. Mitral valve
replacement for mitral regurgitation with and without pres-
ervation of chordae tendineae. J Thorac Cardiovasc Surg
1984;88:718 –25.
14. Calafiore A, Di Mauro M, Gallina S, et al. Mitral valve
surgery for chronic ischemic mitral regurgitation. Ann Tho-
rac Surg 2004;77:1989 –97.
15. Cohn LH, Rizzo RJ, Adams DH, et al. The effect of patho-
physiology on the surgical treatment of ischemic mitral
regurgitation: operative and late risks of repair versus re-
placement. Eur J Cardiothorac Surg 1995;9:568 –74.
16. Magne J, Girerd N, Sénéchal M, et al. Mitral repair versus
replacement for ischemic mitral regurgitation: comparison
of short-term and long-term survival. Circulation 2009;
120(11 Suppl):S104–11.
17. Al-Radi OO, Austin PC, Tu JV, David TE, Yau TM. Mitral
repair versus replacement for ischemic mitral regurgitation.
Ann Thorac Surg 2005;79:1260 –7.
18. McGee EC, Gillinov AM, Blackstone EH, et al. Recurrent mitral
regurgitation after annuloplasty for functional ischemic mitral
regurgitation. J Thorac Cardiovasc Surg 2004;128:916 –24.
19. DeAnda AJ, Komeda M, Nikolic SD, Daughters GT, Ingels
NB, Miller DC. Left ventricular function, twist, and recoil
after mitral valve replacement. Circulation 1995;92:II458 – 66.
20. Ruel M, Kulik A, Rubens FD, et al. Late incidence and
determinants of reoperation in patients with prosthetic heart
valves. Eur J Cardiothorac Surg 2004;25:364 –70.
21. Comparing the Effectiveness of Repairing Versus Replacing
the Heart’s Mitral Valve in People With Severe Chronic
Ischemic Mitral Regurgitation Clinical Trials Registration.
Available at http://clinicaltrials.gov/ct2/show/NCT00807040.
Accessed January 20, 2011.
22. Gillinov AM, Wieryp PN, Blackstone EH, et al. Is repair
preferable to replacement for ischemic mitral regurgitation?
J Thorac Cardiovasc Surg 2001;122:1125– 41.
23. Zoghbi WA, Enriquez-Sarano M, Foster E, et al. Recommen-
dations for evaluation of the severity of native valvular
regurgitation with two-dimensional and Doppler echocardi-
ography. J Am Soc Echocardiogr 2003;16:777– 802.
24. Lang RM, Bierig M, Devereux RB, et al. Recommendations
for chamber quantification: a report from the American
Society of Echocardiography’s Guidelines and Standards
Committee and the Chamber Quantification Writing Group,
developed in conjunction with the European Association of
Echocardiography, a branch of the European Society of
Cardiology. J Am Soc Echocardiogr 2005;18:1440 – 63.
25. Quinones MA, Otto CM, Stoddard M, Waggoner A, Zoghbi
WA. Recommendations for quantification of Doppler echo-
cardiography: a report from the Doppler Quantification Task
Force of the Nomenclature and Standards Committee of the
American Society of Echocardiography. J Am Soc
Echocardiogr 2002;15:167– 84.
26. Carpentier A. Cardiac valve surgery—the “French correc-
tion.” J Thorac Cardiovasc Surg 1983;86:323–37.
27. Akins CW, Miller DC, Turina MI, et al. Guidelines for
reporting mortality and morbidity after cardiac valve inter-
ventions. Ann Thorac Surg 2008;85:1490 –5.
28. Diodato MD, Moon MR, Pasque MK, et al. Repair of isch-
emic mitral regurgitation does not increase mortality or
improve long-term survival in patients undergoing coronary
artery revascularization: A propensity analysis. Ann Thorac
Surg 2004;78:794 –9.
29. Glower DD, Tuttle RH, Shaw LK, Orozco RE, Rankin JS.
Patient survival characteristics after routine mitral valve
repair in ischemic mitral regurgitation. J Thorac Cardiovasc
Surg 2005;129:860 – 8.
30. Gillinov AM, Blackstone EH, Rajeswaran J, et al. Ischemic
versus degenerative mitral regurgitation: does etiology affect
survival? Ann Thorac Surg 2005;80:811–9.
31. Bishay ES, McCarthy PM, Cosgrove DM, et al. Mitral valve
surgery in patients with severe left ventricular dysfunction.
Eur J Cardiothorac Surg 2000;17:213–21.
32. Dahlberg PS, Orszulak TA, Mullany CJ, et al. Late outcome
of mitral valve surgery for patients with coronary artery
disease. Ann Thorac Surg 2003;76:1539 – 48.
33. Grossi EA, Woo YJ, Patel N, et al. Outcomes of coronary
artery bypass grafting and reduction annuloplasty for func-
tional ischemic mitral regurgitation: a prospective multi-
center study (Randomized Evaluation of a Surgical Treat-
ment for Off-Pump Repair of the Mitral Valve). J Thorac
Cardiovasc Surg 2011;141:91–7.
34. Ikonomidis JS, Kratz JM, Crumbley AJ 3d, et al. Twenty-year
experience with the St Jude Medical mechanical valve pros-
thesis. J Thorac Cardiovasc Surg 2003;126:2022–31.
35. Chan V, Kulik A, Tran A, et al. Long-term clinical and hemo-
dynamic performance of the Hancock II versus the Perimount
aortic bioprostheses. Circulation 2010;122(11 Suppl):S10 – 6.
36. Chan V, Jamieson WR, Lam BK, et al. Influence of the On-X
mechanical prosthesis on intermediate-term major throm-
boembolism and hemorrhage: a prospective multicenter
study. J Thorac Cardiovasc Surg 2010;140:1053– 8.
37. Fuster V, Ryden LE, Cannom DS, et al. ACC/AHA/ESC 2006
Guidelines for the Management of Patients With Atrial
Fibrillation A Report of the American College of Cardiology/
American Heart Association Task Force on Practice Guide-
lines and the European Society of Cardiology Committee for
Practice Guidelines (Writing Committee to Revise the 2001
Guidelines for the Management of Patients With Atrial
Fibrillation). J Am Coll Cardiol 2006;48:854 –906.
DISCUSSION
DR ROBERT DION (Genk, Belgium):A very nice presentation.
Of course, nobody can ever prove that replacement is inferior to
repair without a prospective randomized study.
You present 25% recurrence of at least moderate MR at 2
years after valve repair but, you have used, as the Cleveland
Clinic in its original paper by McGee,a variety of repair
techniques and devices: a band, an incomplete band, a
complete ring, . . . You do not precise either how restrictive
the mitral valve annuloplasty was.
And my problem with that type of study with inhomoge-
neous techniques and devices resulting in a 25% recurrence of
moderate MR at 2 years, is that you certainly do not favor the
restrictive mitral annuloplasty! In Leiden we have published
13% of recurrence after 4 years because we use, all of us, the
same strategy. Therefore, my first question: how do you
achieve the restriction of the mitral annulus, what is your
endpoint, do you measure the leaflet coaptation, and when do
you decide to use an incomplete band versus a rigid complete
ring?
And my second question is about the outcome of the patient.
In your series, the mean left ventricular end-diastolic diameter
was 60 mm. In our experience, these LV undergo a reverse
remodeling after a successful and durable restrictive annulo-
plasty. The same should happen after a mitral valve replace-
ment, supposedly by using a bioprosthesis. What happens then
after 5 years in a patient of, let us say, 50 years who is optimally
revascularized and who has ‘reversely remodeled’: a reoperation
for a degenerated bioprothesis in mitral position?
1365Ann Thorac Surg CHAN ET AL
2011;92:1358–66 MITRAL REPAIR VERSUS REPLACEMENT FOR IMR
ADULT CARDIAC
DR CHAN: Thank you for your questions, Dr. Dion.
Your first question relates to the nature of our mitral valve
repair techniques, and also the impact of recurrent 2-plus MR
on outcomes. Seventy-five percent of the patients in this study
received the Medtronic Futureband, which is a rigid band.
The paper that you quoted from the Cleveland Clinic reported
a 28% rate of recurrent MR at 6 months. The focus of that
paper was to compare different annuloplasty bands, some of
which were softer that the rigid Medtronic Futureband. In this
study, few patients received a complete O-ring like the
Physio-ring or a soft Duran band.
At our institution, if excessive tethering of the posterior leaflet
was observed intra-operatively, the Alfieri technique was uti-
lized. Thirty-four patients, or over half of our repair patients,
received an edge-to-edge repair.
The rate of recurrent MR greater than or equal to 2-plus is not
well-defined in the literature. The paper that you quoted from
the Cleveland Clinic addressed recurrent 3- and 4-plus MR. Our
experience is that recurrent moderate MR may not impact
clinical endpoints such as New York Heart Association func-
tional class. This data was not included in today‘s presentation.
Your other question related to mitral valve replacement. Prior
to surgery, the operating surgeon would discuss the advantages
and disadvantages of different mitral valve prostheses as part of
the preoperative consent process. In younger patients, a me-
chanical valve may be preferable. In our series, 17 patients
received the On-X valve, whereas the majority received a tissue
valve. This is a reflection of the patient demographics of this
study. One patient developed premature structural deteriora-
tion of a porcine mitral valve approximately 4 years following
initial mitral surgery. That patient required reoperation for
mitral stenosis rather than insufficiency. These data highlight
the tradeoff between freedom from valve-related complications
versus the rate of recurrent mitral insufficiency, when consider-
ing mitral valve repair versus replacement.
DR SCOTT MCCLURE (London, Ontario, Canada):Vince, good
job as per usual. Just a quick question.
To buy into these results, obviously, as you know, with a study
such as this, we have to believe in the propensity model. And my
question is more to do with your model. If I understood your
study, you were able to match 65 of 69 patients to a pool of only
131. Is that correct?
DR CHAN: Between 2001 and 2010, 230 patients underwent
mitral valve surgery for ischemic mitral regurgitation at our
institution. Of these, 161 underwent mitral valve repair, and 69
patients underwent replacement.
The 65 replacement patients included in this study represent
individuals that presented to the OR with the intention of
receiving a prosthetic mitral valve. This is different from indi-
viduals for which mitral valve repair was attempted, prolonged
cardiopulmonary bypass was required, and eventually, received
mitral valve replacement.
Several aforementioned preoperative variables were used to
calculate a propensity score in order to facilitate matching with
65 mitral valve repair patients.
Another way to analyze these data would have been to use a
regression model. We did repeat the analysis with a propor-
tional hazards model, and similar results were achieved. We
ultimately decided to present the data using a propensity score
in order to simplify our results.
DR MCCLURE: And with regards to the preoperative variables you
used, like ejection fraction, did you simply dichotomize the variable
above and below a particular threshold? How liberal, or put
another way, how strict were you? Because when you look at the
baseline demographics, you “means” were similar for left ventric-
ular diameter as an average across the whole group; but for the
individual matching, how strict were you?
And with regards to the grafts, did you simply dichotomize
the variable: Yes, you had a graft, or no, you didn’t? Or did you
actually model them so you matched individuals patients based
on grafts to areas with an ischemic burden?
Because if we don’t have confidence in the model, then I just
don’t know that you can be as confident in the results.
DR CHAN: I appreciate your comments. Ejection fraction was
categorized into four strata of increasing left ventricle dysfunc-
tion. This is consistent with the outcomes papers published from
our institution. One benefit of this categorization is that it allows
minor measurement and recording errors within each strata.
Concomitant coronary artery bypass grafting was used as a
dichotomous variable for all regression models.
DR ZIVOJIN JONJEV (Sremska Kamenica, Serbia):Repair or
replacement is a very simple question, but a very complex prob-
lem. At the present time we have a lot of information about
Ischemic Mitral Regurgitation. One of the most important things is
the extent of remodeling of the left ventricle and it’s ejection
fraction. You mentioned that you have a significant number of
patients with ejection fraction lower than 35%. In most of the
recently published literature the cutoff point in Ischemic Mitral
Regurgitation is ejection fraction lower than 30%, and those pa-
tients have completely different treatment, prognosis, and survival
curves. So those patients should be observed and presented
separately.
The second issue is decision-making process in your team.
How do you make patient selection for repair or replacement?
Do you respect sphericity index, mitral valve tenting, and
ejection fraction together, or do you have some other criteria?
And finally, your survival and follow up curves. Significant
difference between repair and replacement is detectable 5 or 6
years after surgery. Your follow up lasts just 5 or 6 years. So
maybe you should extend your research in the future for a
longer period of time.
DR CHAN: Thank you kindly for your comments.
Preoperative ejection fraction is an important predictor of
long-term outcomes. The intention of this study was to investi-
gate repair versus replacement in a “real-world” population of
ischemic MR. By matching patients, we had a similar proportion
of LV dysfunction between groups.
In this study, no assessments of preoperative myocardial
viability were made. The extent of myocardial viability would
determine the degree of LV recovery following coronary artery
bypass grafting. As such, preoperative LV size alone may not
completely predict remodeling following surgery.
The decision as to whether or not a valve was repairable was
ultimately made by the operating surgeon. Intra-operative echo-
cardiographic measurements such as posterior angle, posterior
leaflet length, coaptation distance were not presented in this
data today. I would argue that these echocardiographic param-
eters are important for predicting recurrent MR in the early
postoperative period. In this series, the 6-month freedom from
recurrent MR more than 2 plus was 95%. In my opinion,
variables that portend poor durability of mitral valve repair in
the long term are not well described.
And I absolutely agree that these patients should be followed
further into the future. Although we report no difference in
survival between groups, a difference may be observed with
longer follow-up.
1366 CHAN ET AL Ann Thorac Surg
MITRAL REPAIR VERSUS REPLACEMENT FOR IMR 2011;92:1358– 66
ADULT CARDIAC
