Original Article

Asian Cardiovascular & Thoracic Annals

21(3) 288–292
ß The Author(s) 2012
Effect of mitral valve replacement Reprints and permissions:
sagepub.co.uk/journalsPermissions.nav
on reduction of left atrial size DOI: 10.1177/0218492312453142
aan.sagepub.com

Shantanu Pande, Surendra K Agarwal, Satyapriya Mohanty and

Anubhav Bansal

Abstract
Introduction: asymptomatic left atrial enlargement is not uncommon in rheumatic mitral valve disease. We studied the
change in size of the left atrium after mitral valve replacement.
Patients and methods: between January and December 2005, 116 patients underwent isolated mitral valve replace-
ment. Two groups were identified based on left atrial size: group 1 < 60 mm (n = 79) and group 2 > 60 mm (n = 37). The
patients were followed up for 40.4  19.3 months Clinical assessment, preoperative and last postoperative echocardio-
grams were considered for analysis.
Results: the left atrium decreased by 5.84 mm  10.5 in group 1 compared to 20.9 mm  10.64 in group 2 (p = 0.0001).
This correlated with preoperative mitral valve area (p = 0.009), preoperative mitral regurgitation (p = 0.000), and pre-
operative atrial fibrillation (p = 0.022). Linear regression analysis revealed atrial fibrillation (p = 0.001, b1 = 6.006), a high
grade of mitral regurgitation (p = 0.001, b1 = 3.812), and larger size of the left atrium (p = 0.000, b1 = 0.701) predicted
a greater reduction of left atrial size during follow-up. Left atrial size decreased by 28 mm in patients with a preoperative
left atrium >60 mm (75% sensitivity and 100% specificity).
Conclusion: the asymptomatic left atrium reduces in size considerably after mitral valve replacement, and the decrease
is greater in patients with a left atrium >60 mm in size.

Keywords
Heart valve diseases, heart valve prosthesis implantation, mitral valve, heart atria, mitral valve insufficiency

Introduction
Left atrial (LA) enlargement is commonly witnessed
with mitral valve disease. Increased LA size is asso-
ciated with atrial fibrillation (AF) and the risk of
thrombus formation due to stasis of blood.1 Excessive
LA enlargement can lead to bronchial compression or
hoarseness of voice due to compression of the recurrent
laryngeal nerve.2 While operating for correction of
mitral valve disease, extremely enlarged LA are reduced
surgically.3 Some surgical groups recommend routine
LA reduction with only a modest increase in its size
and without any clinical signs of LA enlargement.4,5
Some groups even advocate reducing the size of the
LA when it is >50 mm.6 Surgical reduction of the LA
increases crossclamp and cardiopulmonary bypass
times, and is sometimes associated with excessive post-
operative bleeding.7 There are no standard recommen-
dations for this procedure based on the size of the LA.
This study observed the change in LA size following
mitral valve replacement over 5 years of follow-up.
Patients and methods
This was a retrospective study of patients who under-
went mitral valve replacement between January 2005
and December 2005. During this period, 116 patients
were operated on for isolated mitral valve replacement
Department of Cardiovascular and Thoracic Surgery, Sanjay Gandhi
Postgraduate Institute of Medical Sciences, Lucknow, India
Corresponding author:
Shantanu Pande, MCh, Department of Cardiovascular and Thoracic
Surgery, Sanjay Gandhi Postgraduate Institute of Medical Sciences,
Lucknow 226014, India.
Email: spande@sgpgi.ac.in

Downloaded from aan.sagepub.com at UCSF LIBRARY & CKM on April 4, 2015

Pande et al. 289

Table 1. Demographic and clinical data according to left atrial size in 116 patients undergoing mitral
valve replacement.

Variable Group 1 (n = 79) Group 2 (n = 37) p value

Age (years) 32  12 33  10 0.5

Male 35 13 0.1
Female 44 22 0.1
Mean NYHA class 2.81  0.39 2.90  0.39 0.2
Atrial fibrillation 23 22 0.0001
Follow-up (months) 40.7  19.3 39.7  19.8 0.8
NYHA: New York Heart Association.

Table 2. Preoperative echocardiography data according to left atrial size in 116 patients undergoing mitral
valve replacement.

Variable Group 1 (n = 79) Group 2 (n = 37) p value

LVEDD (mm) 46.68  9.9 56.83  11.43 0.0001

LVESD (mm) 29.54  7.40 37.86  10.52 0.0001
LVEF 60.45%  1.90% 59.86%  4.77% 0.3
LA size (mm) 47.87  7.66 67.46  6.84 0.3
RVSP (mm Hg) 54.03  24.21 51.10  16.55 0.1
LA: left atrial; LVEDD: left ventricular end-diastolic diameter; LVEF: left ventricular ejection fraction; LVESD: left
ventricular end-systolic diameter; RVSP: right ventricular systolic pressure.

in the Sanjay Gandhi Postgraduate Institute of Medical Table 3. Preoperative mitral valve assessment by Doppler
Sciences. Mitral valve disease was of rheumatic etiology echocardiography, according to left atrial size in 116 patients
in all cases. Two groups were classified on the basis of undergoing mitral valve replacement.
LA size: group 1 < 60 mm (n = 79) and group Group 1 Group 2
2 > 60 mm (n = 37). Data were collected from the hos- Variable (n = 79) (n = 37) p value
pital information system and patient case files. The
demographic profile, clinical data, and duration of MVA (cm2) 1.125  4.66 1.288  0.56 0.1
follow-up are given in Table 1. During follow-up, the MV gradient (mm Hg) 16.91  7.68 14.30  5.08 0.09
patients were interviewed in the outpatient clinic, and MR grade 0.005
echocardiography was performed yearly. Patient None 22 5
follow-up until December 2010 was included in this Mild 12 4
study (range, 12–64 months). The data were similar in Moderate 25 9
both groups except for a significantly higher percentage Severe 20 19
of patients in AF in group 2. Left ventricular dimen-
sions were higher in group 2 (Table 2). A significantly MR: mitral regurgitation; MV: mitral valve; MVA: mitral valve area.
higher grade of mitral regurgitation was present in
group 2 (Table 3).
Patients were operated on through a median sternot- A Philips Sonos 5500 and a 3.2-MHz transducer
omy, using cardiopulmonary bypass with aortic and (Philips Medical Systems, Andover, MA, USA) were
bicaval cannulation. Moderate hypothermia of 32 C used for echocardiographic imaging. Routine echocar-
was achieved, and tepid blood antegrade cardioplegia diography was performed using a prescribed protocol.
was used after ascending aortic crossclamping. Mitral The left ventricular dimensions in systole and diastole,
valve replacement was performed through the left left ventricular ejection fraction, right ventricular
atrium, using a bileaflet mechanical prosthesis. systolic pressure, and valve morphology were assessed
De Vega’s annuloplasty was carried out on an empty utilizing two-dimensional imaging and Doppler meas-
beating heart, with caval snugging, using pledgeted 2/0 urements. LA size was measured in parasternal short-
polypropylene sutures. axis view. If the LA was asymmetrically enlarged, the

Downloaded from aan.sagepub.com at UCSF LIBRARY & CKM on April 4, 2015

290
Table 4. Pre- and postoperative echocardiography parameters according to left atrial size.
Group 1
Variable Preoperative Postoperative
LVEDD (mm) 46.68  9.9 44.60  5.40
LVESD (mm) 29.54  7.40 28.7  5.40
LVEF 60.45%  1.90% 59.0%  4.90%
LA size (mm) 47.87  7.66 41.90  8.40
RVSP (mm Hg) 54.03  24.21 37.8  13.0
LA: left atrial; LVEDD: left ventricular end-diastolic diameter; LVEF: left ventricular ejection fraction; LVESD: left ventricular end-systolic diameter;
RVSP: right ventricular systolic pressure.
maximum diameter was recorded. Mitral valve regurgi-
tation was graded on a scale of 1–4. Mitral valve area
and gradient were calculated by continuous-wave
Doppler. The analysis was based on preoperative and
last postoperative echocardiographic and clinical data.
Postoperative events including pleural and pericardial
effusion, complete heart block, severe anemia, pneumo-
thorax, delayed chest closure, and prolonged pacing
were recorded in both groups.
Data are represented as mean  standard deviation.
Comparisons between the 2 groups were made using
Student’s t test and when the sample sizes were
uneven, and small nonparametric test, Mann Whitney
U test, was applied. Comparisons within groups
were undertaken with the Wilcox Anderson test.
Receiver operating characteristic curves were plotted
for factors that correlated significantly with LA size
in the preoperative period and the postoperative
decrease in LA size. Liner regression analysis was
applied for factors correlating significantly with the
decrease in LA size. All analyses were carried out
with SPSS version 10 for Windows (SPSS, Inc.,
Chicago, IL, USA).
Results
There were 2 immediate postoperative deaths, one in
each group, thus follow-up of 114 patients is presented.
Both of the patients who died had septicemia with mul-
tiorgan failure. Immediate postoperative events tended
to be higher in group 2: 4 (5%) in group 1 vs. 6 (16%)
in group 2, p = 0.07. At the last follow-up, 62 of 78
(79%) patients in group 1 were in AF compared to 28
of 36 (78%) in group 2 (p = 0.9). In group 1, 52 of 68
(76%) patients were in NYHA class I compared to 28
of 36 (78%) in group 2. Postoperatively, there was a
reduction in LV size in both groups, but to a greater
extent in group 2. LA size and right ventricular systolic
pressure were significantly reduced in both groups
(Table 4). LA size decreased by 5.84  10.5 mm in
group 1 compared to 20.9  10.64 mm in group 2
Downloaded from aan.sagepub.com at UCSF LIBRARY & CKM on April 4, 2015
Asian Cardiovascular & Thoracic Annals 21(3)
Group 2
p value Preoperative Postoperative p value
0.05 56.83  11.43 45.90  5.80 0.0001
0.3 37.86  10.52 29.50  4.30 0.0001
0.03 59.86%  4.77% 60.60%  2.70% 0.6
0.0001 67.46  6.84 48.20  8.80 0.0001
0.01 51.10  16.55 35.19  6.70 0.01
(p = 0.0001). Preoperative LA size had a strong posi-
tive correlation with preoperative mitral regurgitation
(r = 0.296, p = 0.002), a strong negative correlation
with AF (r = 0.0541, p = 0.000), and very strong
negative correlation with LA size on follow-up
(r = 0.724, p = 0.000). No correlation was detected
between LA size and preoperative mitral valve gradient
or right ventricular systolic pressure. The decrease in
LA size postoperatively correlated with preoperative
mitral valve area (r = 0.26, p = 0.009), preoperative
mitral regurgitation (r = 0.389, p = 0.000), and pre-
operative AF (r = 0.223, p = 0.022). Linear regres-
sion analysis of these factors as predictors of a
decrease in LA size showed AF (p = 0.001,
b1 = 6.006), a high grade of mitral regurgitation
(p = 0.001, b1 = 3.812), and larger LA size
(p = 0.000, b1 = 0.701) predicted a better chance of
a decrease in LA size during follow-up. Preoperative
right ventricular systolic pressure correlated with pre-
operative mitral valve area (r = 0.262, p = 0.009),
preoperative mitral regurgitation (r = 0.223,
p = 0.02), and preoperative mitral valve gradient
(r = 0.536, p = 0.05). The receiver operating character-
istic curve of preoperative AF and LA size was highly
significant (p = 0.000, area under curve = 0.824). LA
size of 41.50 mm was associated with AF (100% sensi-
tivity and 75% specificity). The receiver operating char-
acteristic curve of the decrease in LA size in the 2
groups was significant (p = 0.000, area under
curve = 0.167). LA size decreased by 28 mm in patients
with preoperative LA size > 60 mm with 75% sensitiv-
ity and 100% specificity (Figure 1).
Discussion
This study indicates that enlargement of the LA is
dependent on the presence of AF and mitral regurgita-
tion. There was no effect of mitral valve area or right
ventricular systolic pressure on the size of the LA.
Previous studies have found correlations between LA
size and AF as well as mitral regurgitation.8,9 It has
Pande et al.
Figure 1. Receiver operating characteristic (ROC) curves. (a) ROC curve for atrial fibrillation and preoperative left atrial size.
(b) ROC curve for decrease in left atrial size postoperatively in the 2 groups.
been observed that LA size reduces with the return of
normal sinus rhythm and a decrease in the gradient
across the mitral valve.10 In this study, there was a
reduction in the mitral valve gradient and this correlated
well with the decrease in LA size. Although the patients
remained in AF with an increase in the absolute number
of patients in group 1 with AF, yet there was a decrease
in LA size. In rheumatic disease, a giant left atrium is
more common.11 It is considered to be due to weakness
of the atrial wall.12 However, there is no evidence of
Aschoff bodies in biopsies of LA fibrosis, and chronic
inflammatory infiltrates are readily witnessed.13 It was
also observed that although the gradient across the
mitral valve reduced postoperatively, this did not affect
recovery of normal sinus rhythm, but a correlation has
been reported in other studies.14 This difference can be
attributed to the rheumatic etiology in all our patients.
LA size of 41.50 mm was associated with AF with 100%
sensitivity and 75% specificity in our series.
A decrease in LA size from the preoperative level
was associated with preoperative mitral valve area,
AF, and mitral valve regurgitation. LA size reduced
by 28 mm after mitral valve replacement in patients
with preoperative LA > 60 mm. Linear regression ana-
lysis revealed that AF, a higher grade of mitral regur-
gitation, and larger preoperative LA size predicted a
better chance of reduction of LA size on postoperative
follow-up. Our study found a decrease in LA size after
5 years of follow-up, while others have reported a
reduction in LA size in the immediate postoperative
period, but no further reduction during follow-up.15,16
There were more postoperative events in the group with
larger LA, but this was not significant. AF was similar
in both groups during follow-up.
Some reports have advocated a reduction in the size
of the LA during mitral valve surgery, even in
Downloaded from aan.sagepub.com at UCSF LIBRARY & CKM on April 4, 2015
291
asymptomatic patients, to help regain normal sinus
rhythm.17 Others have found no significant reduction
in LA size, regardless of whether or not plication was
performed.18 It has been observed that LA enlargement
is rarely evidenced in non-rheumatic mitral valve dis-
ease.11 Our study supports the observation that there is
a significant reduction in LA size after mitral valve
replacement.19 Furthermore, we found that the reduc-
tion in LA size was more profound in patients with
larger preoperative LA who were in AF and had high
grades of mitral regurgitation. This regression in LA
size was observed in rheumatic valvular heart disease
with asymptomatic LA enlargement. Atrial fibrillation
was observed even in patients with smaller LA sizes,
and was persistent even with the reduction in LA size
during follow-up. Therefore, the left atrium may not
require reduction during mitral valve replacement in
this group of patients.
Funding
This research received no specific grant from any
funding agency in the public, commercial, or not-for-profit
sectors.
Conflicts of interest statement
None declared.
References
1. Garcı́a-Villarreal OA, Rodrı́guez H, Treviño A, Gouveia
AB and Argüero R. Left atrial reduction and mitral valve
surgery: the ‘‘function anatomic unit’’ concept. Ann Thorac
Surg 2001; 71: 1044–1045.
2. Sugiki H, Murashita T, Yasuda K and Doi H. Novel tech-
nique for volume reduction of giant left atrium: simple and
effective ‘‘spiral resection’’ method. Ann Thorac Surg 2006;
81: 378–380.
292
3. Kawazoe K, Beppu S, Takahara Y, Nakajima N,
Tanaka K, Ichihashi K, et al. Surgical treatment of
giant left atrium combined with mitral valvular dis-
ease. Plication procedure for reduction of compression
to the left ventricle, bronchus and pulmonary paren-
chyma. J Thorac Cardiovasc Surg 1983; 85: 885–892.
4. Badhwar V, Rovin JD, Davenport G, Pruitt JC,
Lazarra RR, Erba G, et al. Left atrial reduction
enhances outcomes of modified maze procedure for
permanent atrial fibrillation during concomitant
mitral surgery. Ann Thorac Surg 2006; 82: 1758–1764.
5. Scherer M, Dzemali O, Aybek T, Wimmer-Greinecker G
and Moritz A. Impact of left atrial size reduction on
chronic atrial fibrillation in mitral valve surgery.
J Heart Valve Dis 2003; 12: 469–474.
6. Winlaw DS, Farnsworth AE, Macdonald PS, Mundy JA
and Spratt PM. Left atrial reduction: the forgotten
Batista. Lancet 1998; 351: 879–880.
7. Kutay V, Kirali K, Ekim H and Yakut C. Effect of giant
left atrium on thromboembolism after mitral valve
replacement. Asian Cardiovasc Thorac Ann 2005; 13:
107–111.
8. Henry WL, Morganroth J, Pearlman AS, Clark CE,
Redwood DR, Itscoitz SB, et al. Relation between echo-
cardiographically determined left atrial size and atrial fib-
rillation. Circulation 1976; 53: 273–279.
9. Le Tourneau T, Messika-Zeitoun D, Russo A, Detaint D,
Topilsky Y, Mahoney DW, et al. Impact of left atrial
volume on clinical outcome in organic mitral regurgita-
tion. J Am Coll Cardiol 2010; 56: 570–578.
10. Reant P, Lafitte S, Jais P, Serri K, Weerasooriya R,
Hocini M, et al. Reverse remodeling of the left cardiac
chambers after catheter ablation after 1 year in a series of
patients with isolated atrial fibrillation. Circulation 2005;
112: 2896–2903.
Downloaded from aan.sagepub.com at UCSF LIBRARY & CKM on April 4, 2015
Asian Cardiovascular & Thoracic Annals 21(3)
11. Hurst JW. Memories of patient with a giant left atrium.
Circulation 2001; 104: 2630–2631.
12. Johnson J, Danielson GK, MacVaugh 3rd H and Joyner
CR. Plication of giant left atrium at operation for severe
mitral regurgitation. Surgery 1967; 61: 118–121.
13. Plaschkes J, Borman J, Merin G and Milwidsky H. Giant
left atrium in rheumatic heart disease: a report of 18 cases
treated by mitral valve replacement. Ann Surg 1971; 174:
194–201.
14. Fan K, Lee KL, Chow WH, Chau E and Lau CP.
Internal cardioversion of chronic atrial fibrillation
during percutaneous mitral commissurotomy: insight
into reversal of chronic stretch-induced atrial remodeling.
Circulation 2002; 105: 2746–2752.
15. Choo S, Park N, Lee S, Kim JW, Song JK, Song H, et al.
Excellent results of fibrillation surgery in the presence of
giant left atrium and mitral valve disease. Eur J
Cardiothorac Surg 2004; 26: 336–341.
16. Hagihara H, Kitamura S, Kawachi K, Morita R,
Taniguchi S, Fukutomi N, et al. Left atrial plication com-
bined with mitral valve surgery in patient with a giant left
atrium. Surg Today 1995; 25: 338–342.
17. Scherer M, Therapidis P, Miskovic A and Moritz A. Left
atrial size reduction improves the sinus rhythm conver-
sion rate after radiofrequency ablation for continuous
atrial fibrillation in patients undergoing concomitant car-
diac surgery. Thorac Cardiovasc Surg 2006; 54: 34–38.
18. Di Eusanio G, Gregorini R, Mazzola A, Clementi G,
Procaccini B, Cavarra F, et al. Giant left atrium and
mitral valve replacement: risk factor analysis. Eur J
Cardiothorac Surg 1998; 2: 151–159.
19. Apostolakis E and Shuhaiber JH. The surgical manage-
ment of giant left atrium. Eur J Cardiothorac Surg 2008;
33: 182–190.