- Email: [email protected]
Tricuspid annulus diameter does not predict the development of tricuspid regurgitation after mitral valve repair for mitral regurgitation due to degenerative diseases
Accepted Manuscript Tricuspid annulus diameter does not predict the development of tricuspid regurgitation after mitral valve repair for mitral regurgitation due to degenerative diseases Tirone E. David, MD, Carolyn M. David, BN, Cedric Manlhiot, PhD PII:
S0022-5223(18)30266-6
DOI:
10.1016/j.jtcvs.2017.12.126
Reference:
YMTC 12520
To appear in:
The Journal of Thoracic and Cardiovascular Surgery
Received Date: 17 March 2017 Revised Date:
13 November 2017
Accepted Date: 3 December 2017
Please cite this article as: David TE, David CM, Manlhiot C, Tricuspid annulus diameter does not predict the development of tricuspid regurgitation after mitral valve repair for mitral regurgitation due to degenerative diseases, The Journal of Thoracic and Cardiovascular Surgery (2018), doi: 10.1016/ j.jtcvs.2017.12.126. This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
ACCEPTED MANUSCRIPT
1
Tricuspid annulus diameter does not predict the development of tricuspid regurgitation
2
after mitral valve repair for mitral regurgitation due to degenerative diseases
RI PT
3
Tirone E. David, MD, Carolyn M. David, BN, and Cedric Manlhiot, PhD
5
From the Division of Cardiovascular Surgery of Peter Munk Cardiac Centre at Toronto
6
General Hospital and University of Toronto, Toronto, Ontario, Canada
SC
4
M AN U
7
8
Read at the 97th Annual Meeting of the American Association for Thoracic Surgery, April
9
29 – May 2, 2017, Boston, MA,
13
14
15
EP
12
Keywords: tricuspid valve, tricuspid annulus, tricuspid insufficiency
Conflict of Interest Statement: The authors have no conflict of interest to disclose
AC C
11
TE D
10
Funding: This project was funded by Miranda and Anthony Wong
16
17
Word count: 4085 (entire manuscript)
1
ACCEPTED MANUSCRIPT
18
19
Address correspondence to: Tirone E. David, MD 200 Elizabeth St. 4N453, Toronto, Ontario, M5G 2C4 – Canada
21
Email: [email protected]
22
Telephone: 416-340-5062
RI PT
20
SC
Fax: 416-340-4020
M AN U
23
Abbreviations:
25
AF = atrial fibrillation
26
BSA = body surface area
27
MA = mitral annulus
28
MV = mitral valve
29
MR = mitral regurgitation
30
NYHA = New York Heart Association
31
TA = tricuspid annulus
32
TR = tricuspid regurgitation
33
TV = tricuspid valve
34
TVA = tricuspid valve annuloplasty
AC C
EP
TE D
24
2
ACCEPTED MANUSCRIPT
Central message – A preoperative tricuspid annulus diameter ≥40 mm was not associated
36
with the development of postoperative functional tricuspid regurgitation after mitral valve
37
repair for degenerative diseases.
RI PT
35
38
39
Clinical Perspective – Heart valve guidelines suggest that tricuspid annuloplasty should
41
be performed during left side valve operations in the absence of tricuspid regurgitation if
42
the diameter of the tricuspid annulus is ≥40 mm. In this observational study, preoperative
43
tricuspid valve diameter was not associated with odds of postoperative tricuspid
44
regurgitation in either the univariable or multivariable regression models.
M AN U
SC
40
Central Picture: Probability of postoperative tricuspid regurgitation after mitral valve
48
repair for degenerative disease in patients with tricuspid annulus ≥40 mm.
49
50
51
EP
47
AC C
46
TE D
45
52
53 3
ACCEPTED MANUSCRIPT
Abstract
55
Objective: Heart valve surgery guidelines suggest that tricuspid annuloplasty may be
56
beneficial in patients with a tricuspid annulus (TA) ≥40 mm even in the absence of
57
functional tricuspid regurgitation (TR) at the time of surgery for left side valve lesions
58
(Class 2a). Given the broad spectrum of degenerative diseases that affect the
59
atrioventricular valves we hypothesize that this measurement may not predict TR after
60
mitral valve (MV) repair.
61
Methods: The diameter of the TA was measured preoperatively in a cohort of 312
62
consecutive patients who had isolated MV repair for degenerative diseases. The mean
63
diameter of the TA was 36 mm (95% confidence interval, CI: 35 – 37). TA ≥40 mm was
64
present in 80 patients. Median echocadiographic follow-up was 6.7 (IQR 5.4-8.4) years,
65
and 100% complete. The main end-point of the study was postoperative TR of moderate or
66
greater degree.
67
Results: Thirty patients had new or persistent TR at one point during the follow-up. The
68
probability of postoperative TR at 7 years was 6.6% (95% CI 4.6 - 9.4) for all patients,
69
6.8% (95% CI 4.6 - 10.4) for TA <40mm, and 6.0% (95% CI 2.9 - 12.2) for TA ≥40 mm.
70
Preoperative TA diameter was not associated with odds of postoperative TR in either the
71
univariable or multivariable regression models. In these analyses preoperative TR was the
72
strongest predictor of postoperative TR.
73
Conclusion: TA≥40 mm did not predict the development of postoperative TR after MV
74
repair for degenerative diseases.
AC C
EP
TE D
M AN U
SC
RI PT
54
4
ACCEPTED MANUSCRIPT
Functional tricuspid regurgitation (TR) is due to leaflet tethering secondary to
76
right ventricular and annular dilatation.1 Right ventricular dilatation is often caused by
77
increased right ventricular pressure and/or volume overload, but it can also be caused by
78
right ventricular infarction or idiopathic cardiomyopathy. Functional TR is common in
79
patients with advanced left sided heart valve diseases. Current guidelines on the
80
management of heart valve diseases2-3 suggest that tricuspid valve annuloplasty (TVA)
81
may be beneficial in patients with trace or mild functional TR and a tricuspid annulus (TA)
82
diameter ≥40 mm by echocardiography or 70 mm by direct inspection of the tricuspid
83
valve (TV) during surgery for left sided valve lesions (Class 2a, which means that it should
84
be considered but the evidence is conflicting). The concept of performing TVA for TA
85
dilatation rather than the TR grade was introduced by Dreyfus and colleagues4 who
86
hypothesized that dilated TA was a permanent and even progressive lesion that if left
87
unattended at the time of mitral valve (MV) repair it may progress and cause severe TR
88
postoperatively. There is evidence that the TA dilates and changes its shape from elliptical
89
to spherical in patients with severe TR.1,5 What remains controversial is whether TVA
90
should be performed or not if the TR diameter measures 40 mm or more.
EP
TE D
M AN U
SC
RI PT
75
Degenerative diseases of the MV is the most common cause of mitral regurgitation
92
(MR) in North America but include a broad spectrum of lesions in the leaflets and annulus
93
of the atrioventricular valves. We hypothesize that a TA diameter ≥40 mm in patients with
94
MR due to degenerative diseases does not predict postoperative TR after MV repair for
95
degenerative diseases. This study examines the effect of the diameter of the TA in the
96
development of TR after MV repair for MR due to degenerative diseases.
97
Patients and methods
AC C
91
5
ACCEPTED MANUSCRIPT
98 99
This study was approved by the Review Ethics Board of University Health Network and individual patient consent was required. All patients (1,247) who had MV repair for MR due to degenerative diseases from January 1985 through December 2010 by
101
one surgeon (TED) have been followed prospectively. Patients operated since October
102
2005 had preoperative TA diameters measured in systole and diastole. This new entry was
103
prompted by the publication by Dreyfus and colleagues on the importance of the
104
preoperative TA size in predicting postoperative TR after MV repair.4 Of 335 patients
105
operated on from October 2005 through December 2010, 25 had concomitant TVA and
106
were excluded from this study. Patients were followed by the referring cardiologists and
107
seen by the surgeon at 2 to 3 months and contacted by his research personnel every 2 to 3
108
years thereafter. Echocardiography was performed preoperatively, during surgery, and
109
postoperatively at one week, and at every follow-up contact unless there was a reason to
110
image the heart more often. A total of 984 postoperative echocardiograms were available
111
for analysis: 303 (97%) patients had more than one study and 87% had more than 2
112
studies. Most postoperative studies were performed at certified echocardiography
113
laboratories and read by an experienced echocardiographer. MR and TR were recorded as
114
none (0), trace (1+), mild (2+), moderate (3+) and severe (4+), according to the guidelines
115
described by the American Society of Echocardiography.6 If the echocardiographer
116
reported the MR or TR as “mild to moderate” or “moderate to severe”, which occurred in
117
27 studies, the recorded images were reviewed by blinded echocardiographers from our
118
institution and entered into the database according to their interpretation as mild (6
119
patients), moderate (19 patients) or severe (2 patients). The diameters of the tricuspid and
120
mitral annuluses were measured preoperatively by transesophageal echocardiography at
AC C
EP
TE D
M AN U
SC
RI PT
100
6
ACCEPTED MANUSCRIPT
end of diastole in the 4-chamber view at 0º for the TA and at 0º and 90º for the mitral
122
annulus (MA). Adverse events were recorded according to guidelines set by cardiac
123
surgical societies.7 The cause of death was determined by hospital charts review, death
124
certificates, or information from the physician who was caring for the patient at that time.
125
Clinical and echocardiographic follow-up was 100% complete; the median (IQR) duration
126
of clinical follow-up was 7.1 (5.8-8.8) years and the median (IQR) duration of
127
echocardiographic follow-up was 6.7 (5.4-8.4) years. There were no substantial differences
128
between the number of patients reaching landmark durations of follow-up when comparing
129
clinical and imaging follow-up.
130
Statistical analysis
M AN U
SC
RI PT
121
Data is presented as median with interquartile range (IQR, 25th and 75th percentiles)
132
and frequencies as appropriate. Pearson correlation was used to determine the association
133
between preoperative TA and MA size along with linear regression models. Non-linearity
134
of the association between preoperative TA and MA size was assessed through various
135
mathematical transformations of MA size; the Akaike Information Criteria (AIC) was used
136
to compare the different models. Linear regression was used to determine the association
137
between preoperative TA, preoperative TR and degree of myxomatous degeneration
138
(separately, both modelled as categorical variables). Kaplan-Meier method was used to
139
estimate survival over time. Cumulative incidence of stroke and reoperation were
140
calculated with death as a competing risk. Given the low number of patients experiencing
141
any of those 3 outcomes, no attempts at identifying risk factors was performed. The
142
progression of moderate or severe TR over time was modelled using logistic regression
143
models adjusted for repeated measures through an autoregressive covariance structure
AC C
EP
TE D
131
7
ACCEPTED MANUSCRIPT
(mixed longitudinal regression models). The autoregressive covariance structure was chosen
145
specifically because this structure assumes homogenous variance over time and correlations
146
between the different observations on the same patient that decline exponentially over time. Both
147
assumptions are true in this type of data (repeated echocardiograms over time). Univariable risk
148
factor analysis was performed using this approach. Given the limited number of outcomes available
149
for analysis, we elected to perform multivariable risk factor analysis using a priori selection of
150
variables based on clinical relevance. Variables selected as potential risk factors were: sex, age at
151
the time of surgery, preoperative atrial fibrillation/concomitant Maze procedure, preoperative
152
hypertension, preoperative TA ≥40 mm, preoperative TR grade, and time since index procedure.
153
The association between TA size ≥40 mm and progression of moderate or severe TR over
154
time was assessed for various subgroup of patients to determine whether there was any
155
interaction between clinical characteristics and the effect of preoperative TA size on
156
postoperative TR.
SC
M AN U
TE D
157
RI PT
144
Missing data was rare (2.0%) and randomly distributed, in consequence mean imputation was used to handle the few missing data elements; there was no missing outcomes
159
data. All statistical analyses were performed using SAS v9.4 (SAS statistical software, Cary NC).
160
Results
161
Patient profile and operative data for the entire cohort and stratified by preoperative TA
162
size are shown in Tables 1 and 2 respectively.
163
Clinical outcomes - There were 2 operative and 11 late deaths: 2 due to stroke, 1
164
endocarditis and 10 non-cardiac deaths. There was one MV re-repair for recurrent MR 5
165
years after surgery and one MV re-repair for mitral stenosis caused by pannus 7 years after
AC C
EP
158
8
ACCEPTED MANUSCRIPT
surgery. Six patients suffered a stroke (2 died). Table 3 shows the freedom from morbid
167
events over time.
168
Tricuspid and mitral valves – There was a modest positive linear association between
169
preoperative TA and MA diameters as shown in Figure 1 and between TA diameter and
170
TR grade as shown in Figure 2, but no association between TA diameter and the degree of
171
degeneration of the MV as seen in Figure 3. Eighty patients (26%) had TA ≥40 mm before
172
surgery. Thirty patients had new or persistent TR ≥3+ at one point during the follow-up.
173
Figure 4 shows the probability of postoperative TR≥3+ during the follow-up in all patients,
174
in patients with TA≥40 mm, and in patients with TA <40 mm separately. Table 4 shows
175
the results of univariable model for variables associated with postoperative TR≥3+ and
176
Table 5 shows the results of multivariable model. Preoperative TA diameter was not
177
associated with odds of postoperative TR in either the univariable or multivariable
178
regression models. This remained true regardless of whether TA size was modelled as a
179
continuous variable, an ordinal variable or using the threshold of ≥40 mm. There was a
180
significant association between TA size indexed to body surface area and odds of
181
postoperative TR in univariable analysis (Table 4), however, this association did not
182
remain significant in the multivariable regression model (Table 5). In subgroup analysis
183
(Table 6), preoperative TA diameter ≥40 mm was not associated with increased risk of
184
postoperative TR in any subgroup of patient except for patients with preoperative ejection
185
fraction <60%.
186 187
AC C
EP
TE D
M AN U
SC
RI PT
166
Twelve patients developed MR ≥3 and one patient developed mitral stenosis due to pannus during the follow-up.
9
ACCEPTED MANUSCRIPT
188
189
Discussion The concept of performing TVA based on the diameter of the TA rather than the severity of TR was introduced by Dreyfus and colleagues4 in 2005. These investigators
191
systematically explored the TV at the time of MV repair for MR due to various pathologies
192
in 311 patients and performed TVA whenever the distance from the antero-septal
193
commissure to the antero-posterior commissure was ≥70 mm.4 They showed that this
194
approach greatly reduced the risk of developing TR after MV repair.4 We are not aware of
195
any study that correlated this surgical measurement in the arrested heart with preoperative
196
TA diameter, but 40 mm appeared in the guidelines.2-3 Although we do not dispute that a
197
dilated TA is often associated with functional TR, the value 40 mm as a cut off to
198
recommend TVA without TR at the time of MV repair for degenerative MR needs further
199
evaluation. The TV leaflets like the MV leaflets vary in size in patients with degenerative
200
diseases. Patients with fibroelastic deficiency frequently have small MV leaflets, so small
201
that at one time we were reluctant to use an annuloplasty ring or band from fear of creating
202
mitral stenosis.8 The TV leaflets are also small in these patients and a TA of 40 mm is
203
probably too large and TVA is probably necessary to prevent late TR. On the other hand,
204
patients with myxomatous degeneration of the MV frequently have large leaflets with large
205
MA and the TV leaflets are also large with consequent larger annuluses. This type of
206
degenerative disease is present in most of our patients and a TA of 40 mm is probably
207
normal and seldom are associated with severe TR (at least in our experience), and when
208
they are, it is often because of leaflet prolapse rather than annular dilatation. In this study,
209
the diameter of the TA correlated modestly with the size of the MA (Figure 1), as well as
210
with the severity of preoperative TR (Figure 2) but not with the degree of myxomatous
AC C
EP
TE D
M AN U
SC
RI PT
190
10
ACCEPTED MANUSCRIPT
degeneration, which surprised us because we expected that patients with more advanced
212
degrees of myxomatous degeneration would have large TV leaflets and consequently
213
larger TA.
214
RI PT
211
TA ≥40 mm or TA ≥21 mm/m2 did not predict the development of postoperative TR. Preoperative TA diameter was not associated with postoperative TR in either the
216
univariable or multivariable regression models. This remained true regardless of whether
217
TA diameter was modelled as a continuous variable, an ordinal variable or using the
218
threshold of ≥40 mm. There was a significant association between TA size indexed to body
219
surface area and odds of postoperative TR in univariable analysis, however, this
220
association did not remain significant in the multivariable regression model. Preoperative
221
TA diameter ≥40 mm was not associated with increased risk of postoperative TR in any
222
subgroup of patient except for patients with preoperative ejection fraction <60%. In a
223
recent publication, we examined a much larger sample of patients and found that impaired
224
left ventricular function was associated with the development of postoperative TR.9 The
225
most powerful predictor of postoperative TR was the presence of untreated moderate TR
226
before surgery.
M AN U
TE D
EP
We are not the first ones to challenge the concept that TA≥40 mm predicts
AC C
227
SC
215
228
postoperative functional TR after MV repair for degenerative diseases. Sordelli and
229
colleagues10 from Milan, Italy, prospectively evaluated the predictive value of TA diameter
230
assessed by 3D transesophageal echocardiography (3D-TEE) in the development of TR in
231
706 patients who had isolated MV repair for MR due to degenerative diseases (77%
232
myxomatous – with its broad spectrum of lesions – and 23% fibroelastic deficiency).
233
Those investigators performed a detailed analysis of the diastolic and systolic TA 11
ACCEPTED MANUSCRIPT
diameters (antero-posterior and septo-lateral) and the outcomes of surgery after a mean
235
follow-up of 2 years (range 0.5 to 5 years). TR decreased by one grade in 32% of patients,
236
remained unchanged in 62% and increased by one grade in 5.5% but only 3 patients
237
developed moderate or severe TR. The authors concluded that “newly developed
238
significant TR is a rare event after successful repair of degenerative MR” and that
239
“analysis of TA does not predict early to midterm subsequent TR progression”. The MV
240
pathology (myxomatous vs fibroelastic deficiency) was not entered into the multivariable
241
analysis models in that study. It is noteworthy mentioning that the sizes of the TA and MA
242
in our study were very similar to those reported by Sordelli and colleagues.10
M AN U
SC
RI PT
234
A recent report based on the STS database on TVA at the time of MV surgery
244
indicated that concomitant TVA was performed in 14.3% of all MV operations but in
245
centers that performed more than 10 MV operations per year it ranged from 75% in 2
246
centers to less than 10% in more than 300 centers.11 In addition, TVA was performed in
247
only 3.5% when the grade of TR was mild or less, 30.6% when it was moderate and 75.6%
248
when severe.11 However, as documented in this study and a previous one9 preoperative
249
moderate TR is associated the postoperative TR of moderate or severe degree after MV
250
repair for degenerative diseases and we now bow believe that TVA is appropriate in these
251
patients. In addition, TVA should be considered in older patients with longstanding atrial
252
fibrillation, dilated atria, impaired left ventricular systolic function probably should have
253
TVA, particularly if they are women even in absence of moderate or severe TR.9, 12
254
Limitations of study – This is a retrospective study and the TA and MA diameters were
255
obtained intraoperatively and the values may have been affected by the anesthetic, assisted
256
ventilation, and loading conditions which were not controlled at the time of the
AC C
EP
TE D
243
12
ACCEPTED MANUSCRIPT
measurements. The size and area of the TV leaflets were not measured to correlate with the
258
diameter of TA and development of TR. In addition, TR after MV repair for degenerative
259
diseases is uncommon and the sample size of this study was too small and the duration of
260
follow-up too short for more detailed analysis of the variables associated with
261
postoperative TR. Finally, given the limited sample size and the close to significant
262
association between preoperative TA diameter as a continuous variable and increase odds
263
of postoperative TR over time we cannot exclude that the lack of association is in fact a
264
type II error resulting from an underpowered comparison.
265
Conclusions – TA≥40 mm was not associated with the development of TR after MV repair
266
for degenerative diseases, likely because of variable sizes of the TV leaflets in patients
267
with degenerative diseases. Preoperative moderate TR was the most powerful predictor of
268
postoperative TR by multivariable analysis.
271
272
273
SC
M AN U
TE D EP
270
AC C
269
RI PT
257
274
275
276 13
ACCEPTED MANUSCRIPT
277
278
References 1. Sagie A, Schwammenthal E, Padial LR, Vazquez de Prada JA, Weyman AE, Levine RA. Determinants of functional tricuspid regurgitation in incomplete tricuspid valve closure:
280
Doppler color flow study of 109 patients. J Am Coll Cardiol. 1994;24:446-53.
281
RI PT
279
2. Alec Vahanian A, Alfieri O, Andreotti F, Antunes MJ, Barón-Esquivia G, Baumgartner H., et al. Guidelines on the management of valvular heart disease (version 2012) - Eur J
283
Cardio-Thorac Surg 42 (2012) S1–S44.
3. Nishimura RA, Otto CM, Bonow RO, Carabello BA, Erwin III JP, Guyton RA, et al. 2014
M AN U
284
SC
282
285
AHA/ACC Guideline for the management of patients with valvular heart disease:
286
Executive Summary. J Am Coll Cardiol 2014;63:2438-88.
287
4. Dreyfus GD, Corbi PJ, Chan KMJ, and Bahrami T. Secondary Tricuspid Regurgitation or Dilatation: Which Should Be the Criteria for Surgical Repair? Ann Thorac Surg
289
2005;79:127-32.
290
TE D
288
5. Nemoto N, Lesser JR, Pedersen WR, Sorajja P, Spinner E, Garberich RF et al. Pathogenic structural heart changes in early tricuspid regurgitation. J Thorac Cardiovasc Surg
292
2015;150:323-30.
6. Zoghbi WA, Enriquez-Sarano M, Foster E, Grayburn PA, Kraft CD, Levine RA, et al.
AC C
293
EP
291
294
American Society of Echocardiography. Recommendations for evaluation of the severity
295
of native valvular regurgitation with two-dimensional and Doppler echocardiography. J
296
Am Soc Echocardiogr 2003;16:777-802.
297
7. Akins CW, Miller DC, Turina MI, Kouchoukos NT, Blackstone EH, Grunkemeier GL, et
298
al. Guidelines for reporting mortality and morbidity after cardiac valve interventions. J
299
Thorac Cardiovasc Surg 2008;135:732-738. 14
ACCEPTED MANUSCRIPT
301 302 303 304
8. David TE, Armstrong S, McCrindle BW, Manhiolt C. Late outcomes of mitral valve repair for mitral regurgitation due to degenerative disease. Circulation 2013;127:1485-92. 9. David TE, David CM, Fan CPS, Manhiolt C. Tricuspid regurgitation is uncommon after mitral valve repair for degenerative diseases. J Thorac Cardiovasc Surg 2017;154:110-22.
RI PT
300
10. Sordelli C, Lancellotti P, Carlomagno G, Di Giannuario G, Alati E, De Bonis M, et al. Tricuspid Annular Size and Regurgitation Progression After Surgical Repair for
306
Degenerative Mitral Regurgitation. Am J Cardiol 2016;118:424-31
307
SC
305
11. Badhwar V, Rankin S, He M, Jacobs JP, Furnary AP, Fazzalari FL et al. Performing concomitant tricuspid valve repair at the time of mitral valve operations is not associated
309
with increased operative mortality. Ann Thorac Surg 2017;103:587-593.
310
M AN U
308
12. Ro SK, Kim JB, Jung SH, Choo SJ, Chung CH, Lee JW. Mild-to-moderate functional tricuspid regurgitation in patients undergoing mitral valve surgery. J Thorac Cardiovasc
312
Surg 2013;146:1092-7.
315 316
317
318
EP
314
AC C
313
TE D
311
319
320 15
ACCEPTED MANUSCRIPT
Legends
322
Figure 1: Correlation of preoperative tricuspid annulus size vs. preoperative mitral
323
annulus size
324
Figure 2: Tricuspid annulus size vs. preoperative tricuspid regurgitation grade
325
Figure 3: Tricuspid annulus size vs. degree of degeneration of the mitral valve
326
Figure 4 - Estimated punctual probability of postoperative tricuspid regurgitation of
327
moderate degree or greater over time (figure) and at specific follow-up landmarks
328
(embedded table). Data provided for in all patients (upper panel) and in patients with TA
329
<40 mm TA≥40 mm
M AN U
SC
RI PT
321
333
334
335
336
EP
332
AC C
331
TE D
330
337
338 16
ACCEPTED MANUSCRIPT
339 340 341 342
Table 1: Patients profile stratified by preoperative tricuspid annulus diameter
343
Median age in years [IQR]
344
Male sex
345
Body surface area in m2 [IQR]
1.91 [1.75-2.08]
1.90 [1.72-2.06] 2.00 [1.85-2.11]
<0.001
346
Body mass index in kg/m2 [IQR]
25.9 [23.6-28.4]
25.8 [23.2-28.4] 26.3 [24.2-28.6]
0.42
347
Pre-operative heart rhythm
Entire cohort) (N=312) 58 [50-67]
Atrial fibrillation or flutter
44 (14.1)
349
History of atrial fibrillation
350
Complete heart block/pacemaker
59 [49-67]
57 [50-66]
151 (65.1)
P value
0.89
70 (87.5)
<0.001
25 (10.8)
19 (23.8)
0.008
10 (3.2)
7 (3.0)
3 (3.8)
0.75
1 (0.3)
0 (0.0)
1 (1.3)
0.26
M AN U
348
TA ≥40mm (N=80)
SC
221 (70.8)
TA <40mm (N=232)
RI PT
Variable
Previous heart operations
352
Coronary artery bypass
2 (0.6)
2 (0.9)
0 (0.0)
1.00
353
Percutaneous coronary interventions 2 (0.6)
2 (0.9)
0 (0.0)
1.00
354
Atrial septal defect closure
2 (0.9)
0 (0.0)
1.00
355
NYHA functional classification
357
Class II
358
Class III
359
Class IV
360
2 (0.6)
0.24
93 (29.8)
72 (31.0)
21 (26.3)
124 (39.7)
96 (41.4)
28 (35.0)
91 (29.2)
62 (26.7)
29 (36.3)
4 (1.3)
2 (0.9)
2 (2.5)
EP
Class I
AC C
356
TE D
351
Left ventricular ejection fraction
0.08
361
≥ 60%
258 (82.7)
198 (85.3)
60 (75.0)
362
40-59%
46 (14.7)
28 (12.1)
18 (22.5
363
20-39%
8 (2.6)
6 (2.6)
2 (2.5)
108 (34.6)
76 (32.8)
32 (40.0)
364 365
Symptomatology Congestive heart failure
17
028
ACCEPTED MANUSCRIPT
Angina pectoris
367
Syncopal episodes
13 (4.2)
13 (5.6)
0 (0.0)
0.03
4 (1.3)
4 (1.7)
0 (0.0)
0.58
23 (28.8)
368
Associated conditions
369
Hyperlipidemia
111 (35.6)
88 (37.9)
370
Family history of CAD
135 (43.3)
106 (45.7)
371
Diabetes mellitus
15 (4.8)
12 (5.2)
372
Hypertension
115 (36.9)
87 (37.5)
373
Smoking history
124 (39.7)
90 (38.8)
374
Severe COPD
5 (1.6)
375
Previous stroke or TIA
376 377
0.18
RI PT
366
0.15
3 (3.8)
0.77
28 (35.0)
0.79
34 (42.5)
0.60
4 (1.7)
1 (1.3)
1.00
9 (2.9)
9 (3.9)
0 (0.0)
0.12
Infective endocarditis
14 (4.5)
12 (5.2)
2 (2.5)
0.53
Coronary artery disease
43 (13.8)
31 (13.4)
12 (15.0)
0.71
M AN U
SC
29 (36.3)
378 379 Numbers in parentheses are percentages.
381 382
Abbreviatons: IQR = interquartile range; NYHA = New York Heart Association; CAD = coronary artery disease; COPD = chronic obstructive pulmonary disease; TIA = transient ischemic episode
385 386 387 388
EP
384
AC C
383
TE D
380
389 390 391
18
ACCEPTED MANUSCRIPT
392
Table 2. Operative data stratified by preoperative tricuspid annulus diameter
393 394 395
Variable
Entire cohort) (N=312)
TA <40mm (N=232)
Mitral valve prolapse 11 (4.7)
397
Anterior leaflet
21 (6.7)
14 (6.0)
398
Posterior leaflet
112 (35.9)
84 (36.2)
399
Both leaflets
167 (53.5)
123 (53.0)
Echocardiographic measurements: Mitral annulus at 0° in mm [IQR]
40 [37-45]
402
Mitral annulus at 90° in mm [IQR]
40 [37-45]
403
Tricuspid annulus at 0° in mm [IQR]
35 [32-40]
Tricuspid regurgitation grades:
40 [36-43]
7 (8.8)
28 (35.0)
44 (55.0)
43 [39-47]
<0.001
40 [36-44]
42 [39-47]
0.02
34 [32-36]
41 [40-42]
<0.001
M AN U
401
1 (1.3)
RI PT
12 (3.9)
SC
None (annular dilatation)
404
0.06
None or trace (0 – 1+)
150 (48.1)
120 (51.8)
30 (37.5)
406
Mild (2+)
144 (46.2)
101 (43.5)
43 (53.8)
407
Moderate (3+)
18 (5.8)
11 (4.7)
7 (8.8)
TE D
405
Degree of mitral valve myxomatous degeneration* Mild
410
Moderate
411
Advanced
AC C
409
EP
408
P value 0.46
396
400
TA ≥40mm (N=80)
412
Dystrophic calcification of the annulus
413
Operation performed:
0.94
105 (33.7)
79 (34.1)
26 (32.5)
135 (43.3)
99 (42.7)
36 (45.0)
72 (23.1)
54 (23.3)
19 (22.5)
12 (3.9)
10 (4.3)
2 (2.5)
0.74
414
MV repair with Simplici-T band
312 (100)
232 (100)
80 (100)
1.00
415
Chordal replacement
252 (80.8)
184 (79.3)
68 (85.0)
0.32
416
Reconstruction of the mitral annulus
12 (3.9)
10 (4.3)
2 (2.5)
0.74
417
19
ACCEPTED MANUSCRIPT
Maze for atrial fibrillation
54 (17.3)
32 (13.8)
22 (27.5)
0.01
419
Repair of atrial septal defect
15 (4.8)
9 (3.9)
6 (7.5)
0.23
420
Replacement of the ascending aorta
7 (2.2)
6 (2.6)
1 (1.3)
0.68
421
Coronary artery bypass
43 (13.8)
31 (13.4)
12 (15.0)
0.71
422
Resection of atrial myxoma
1 (0.3)
1 (0.4)
423
Septal myectomy
2 (0.6)
1 (0.4)
66 [47-81]
63 [46-100]
425
.
426
Cardiopulmonary bypass time
427
Median [IQR], minutes
Median [IQR], minutes
82 [64-101]
428 Numbers in parentheses are percentages.
430
* Degree of myxomatous degeneration from reference #8
431
Abbreviatons: IQR = interquartile range
434 435 436 437 438 439
EP
433
AC C
432
TE D
429
440 441 442 20
0 (0.0)
1.00
1 (1.3)
0.45
SC
Aortic clamping time
M AN U
424
RI PT
418
81 [62-100]
72 [53-84]
0.06
92 [71-106]
0.04
ACCEPTED MANUSCRIPT
443 444 445
Table 3: Clinical Outcomes - Freedom from mortality from Kaplan-Meier estimates, cumulative incidence of reoperation and stroke over time with death as a competing risk and number of patients with data available at different follow-up landmarks (95% confidence interval, CI)
446
Outcomes
1 year (95% CI)
447
Survival
99.4 (97.5-99.8)
98.4 (96.1-99.3)
448
Reoperation
0.0
449
Stroke
450 451
90.8 (74.2-96.9)
0.7 (0.2-2.6)
0.7 (0.2-2.6)
0.7 (0.2-2.6)
0.3 (0.1-2.3)
0.5 (0.2-2.5)
1.3 (0.5-3.5)
2.9 (1.1-7.3)
N (clinical)
310
297
277
N (imaging)
301
289
255
M AN U
452 453 454 455 456
461 462 463 464 465
EP
460
AC C
459
TE D
457 458
RI PT
97.7 (95.2-98.9)
SC
3 years (95% CI) 5 years (95% CI) 10 years (95% CI)
466 467 468
21
16
10
ACCEPTED MANUSCRIPT
Table 4: Univariable risk factors analysis for postoperative TR ≥3+ over time (odds ratio, OR with 95% confidence interval, CI)
471
Variable
OR
lower CI
upper CI
P-value
472
Age (5-year increment)
1.61
1.34
1.93
<0.001
473
Time since operation (years)
1.05
0.96
1.14
0.27
474
Preoperative MA size (mm)
0.99
0.92
1.06
0.72
475
Preoperative MV area in 0 degrees
0.99
0.93
1.06
0.85
476
Preoperative MV area in 90 degrees 0.96
0.89
1.03
0.22
477
Sex: female
4.08
1.84
9.05
<0.001
478
Atrial fibrillation/Maze procedure
4.63
2.10
10.24
<0.001
479
Hypertension
2.90
1.29
6.55
0.01
480
Preoperative TA size (mm)
1.06
0.99
1.15
0.10
0.88
6.17
0.09
1.74
0.57
5.31
0.33
1.68
0.20
14.02
0.63
3.71
0.44
31.18
0.23
M AN U
SC
RI PT
469 470
<35 mm
reference
482
35-39 mm
2.33
483
40-42 mm
484
42-45 mm
485
>45mm
TE D
481
Preoperative TA size ≥ 40 mm
1.13
0.48
2.68
0.77
487
TA size / BSA > 21 mm/m2
2.416
1.279
4.565
0.007
488
Preoperative TR ≥3+
13.6
5.71
32.59
<0.001
489
Preoperative degeneration of MV (ref.: mild)
490
Moderate
0.86
0.36
2.06
0.74
491
Severe
0.39
0.14
1.07
0.07
492
Preoperative LV grade (per grade)
AC C
EP
486
1.03
0.49
22
2.17
0.94
ACCEPTED MANUSCRIPT
493 494 495
Table 5: Multivariable risk factors analysis for postoperative TR ≥3+ over time for preoperative TA size ≥40mm and preoperative TA size indexed to body surface area (odds ratio, OR with 95% confidence interval, CI)
496
Variable
497
Preoperative TA size ≥40 mm
498
Time since operation (per year)
1.094
0.986
1.213
0.09
499
Age (5-year increment)
1.312
1.054
1.633
0.02
500
Sex (female)
4.244
1.748
10.307
0.001
501
Atrial fibrillation/Maze procedure
1.744
0.666
4.567
0.26
502
Hypertension
2.485
0.958
503
Preoperative TR ≥3+
7.629
2.105
504
Preoperative TA size ≥ 40 mm
1.405
0.492
505
P-value
RI PT
upper CL
SC
lower CL
6.449
0.06
27.658
0.002
4.014
0.53
M AN U
OR
Preoperative TA indexed to body surface area
507
Time since operation (per year)
1.094
0.986
1.213
0.09
508
Age (5-year increment)
1.294
1.050
1.594
0.02
509
Sex (female)
3.735
1.445
9.652
0.007
510
Atrial fibrillation/Maze procedure
1.849
0.680
5.024
0.23
511
Hypertension
2.513
0.965
6.546
0.06
512
Preoperative TR ≥3+
7.537
2.141
26.528
0.002
513
Preoperative TA indexed to BSA
1.024
0.905
1.160
0.70
516 517 518 519
EP
515
AC C
514
TE D
506
520 521 522
23
ACCEPTED MANUSCRIPT
523 524
Table 6: Association between preoperative tricuspid annulus size ≥40mm and postoperative TR ≥3+ over time for various sub-groups of patients (odds ratio, OR with 95% confidence interval, CI)
525
Patient subgroup
526
Age (years)
Lower CL
Upper CL
<50 years
3.376
0.214
53.372
528
50-59 years
0.908
0.750
10.930
529
60-69 years
0.626
0.130
3.011
530
≥70 years
2.013
0.605
6.701
531
Sex
0.39
0.94
0.56
0.25
SC
527
P-value
RI PT
OR
Male
1.191
0.366
3.876
0.77
533
Female
3.210
0.883
11.666
0.08
534
Atrial fibrillation/Maze procedure*
535
Yes
1.472
536
No
0.474
537
Body surface area (m2)
M AN U
532
0.474
4.573
0.50
0.102
2.207
0.34
0.399
9.599
0.40
≤1.75
1.993
539
1.76-1.90
1.052
0.238
4.663
0.95
540
1.91-2.05
1.246
0.114
13.591
0.86
541
>2.05
2.158
0.403
11.547
0.37
1.855
0.310
11.115
0.50
1.518
0.449
5.135
0.50
0.229
0.045
1.167
0.08
Preoperative grade of TR 1-2
544
2
545
≥3+
546
AC C
543
EP
542
TE D
538
Preoperative degeneration of MV
547
Mild
1.540
0.406
5.847
0.53
548
Moderate
0.777
0.200
3.017
0.72
549
Severe
1.642
0.310
8.695
0.56
550
Ejection fraction (%)
551
≥60%
0.672
0.242
1.860
0.44
552
<60%
7.112
1.205
41.980
0.03
24
ACCEPTED MANUSCRIPT
553
Preoperative MA size (mm) ≤37 mm
1.713
0.410
7.151
0.46
555
38-40 mm
2.536
0.257
25.013
0.43
556
41-45 mm
0.264
0.029
2.390
0.24
557
≥46 mm
1.626
0.262
10.091
558
0.60
*Given the small number of patients, the 41 patients with AF and Maze procedure, 13 patients with history of AF
EP
TE D
M AN U
SC
and Maze procedure and the 3 patients with AF but no Maze procedures were combined in a single category.
AC C
559 560
RI PT
554
25
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
S0022-5223(18)30266-6
DOI:
10.1016/j.jtcvs.2017.12.126
Reference:
YMTC 12520
To appear in:
The Journal of Thoracic and Cardiovascular Surgery
Received Date: 17 March 2017 Revised Date:
13 November 2017
Accepted Date: 3 December 2017
Please cite this article as: David TE, David CM, Manlhiot C, Tricuspid annulus diameter does not predict the development of tricuspid regurgitation after mitral valve repair for mitral regurgitation due to degenerative diseases, The Journal of Thoracic and Cardiovascular Surgery (2018), doi: 10.1016/ j.jtcvs.2017.12.126. This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
ACCEPTED MANUSCRIPT
1
Tricuspid annulus diameter does not predict the development of tricuspid regurgitation
2
after mitral valve repair for mitral regurgitation due to degenerative diseases
RI PT
3
Tirone E. David, MD, Carolyn M. David, BN, and Cedric Manlhiot, PhD
5
From the Division of Cardiovascular Surgery of Peter Munk Cardiac Centre at Toronto
6
General Hospital and University of Toronto, Toronto, Ontario, Canada
SC
4
M AN U
7
8
Read at the 97th Annual Meeting of the American Association for Thoracic Surgery, April
9
29 – May 2, 2017, Boston, MA,
13
14
15
EP
12
Keywords: tricuspid valve, tricuspid annulus, tricuspid insufficiency
Conflict of Interest Statement: The authors have no conflict of interest to disclose
AC C
11
TE D
10
Funding: This project was funded by Miranda and Anthony Wong
16
17
Word count: 4085 (entire manuscript)
1
ACCEPTED MANUSCRIPT
18
19
Address correspondence to: Tirone E. David, MD 200 Elizabeth St. 4N453, Toronto, Ontario, M5G 2C4 – Canada
21
Email: [email protected]
22
Telephone: 416-340-5062
RI PT
20
SC
Fax: 416-340-4020
M AN U
23
Abbreviations:
25
AF = atrial fibrillation
26
BSA = body surface area
27
MA = mitral annulus
28
MV = mitral valve
29
MR = mitral regurgitation
30
NYHA = New York Heart Association
31
TA = tricuspid annulus
32
TR = tricuspid regurgitation
33
TV = tricuspid valve
34
TVA = tricuspid valve annuloplasty
AC C
EP
TE D
24
2
ACCEPTED MANUSCRIPT
Central message – A preoperative tricuspid annulus diameter ≥40 mm was not associated
36
with the development of postoperative functional tricuspid regurgitation after mitral valve
37
repair for degenerative diseases.
RI PT
35
38
39
Clinical Perspective – Heart valve guidelines suggest that tricuspid annuloplasty should
41
be performed during left side valve operations in the absence of tricuspid regurgitation if
42
the diameter of the tricuspid annulus is ≥40 mm. In this observational study, preoperative
43
tricuspid valve diameter was not associated with odds of postoperative tricuspid
44
regurgitation in either the univariable or multivariable regression models.
M AN U
SC
40
Central Picture: Probability of postoperative tricuspid regurgitation after mitral valve
48
repair for degenerative disease in patients with tricuspid annulus ≥40 mm.
49
50
51
EP
47
AC C
46
TE D
45
52
53 3
ACCEPTED MANUSCRIPT
Abstract
55
Objective: Heart valve surgery guidelines suggest that tricuspid annuloplasty may be
56
beneficial in patients with a tricuspid annulus (TA) ≥40 mm even in the absence of
57
functional tricuspid regurgitation (TR) at the time of surgery for left side valve lesions
58
(Class 2a). Given the broad spectrum of degenerative diseases that affect the
59
atrioventricular valves we hypothesize that this measurement may not predict TR after
60
mitral valve (MV) repair.
61
Methods: The diameter of the TA was measured preoperatively in a cohort of 312
62
consecutive patients who had isolated MV repair for degenerative diseases. The mean
63
diameter of the TA was 36 mm (95% confidence interval, CI: 35 – 37). TA ≥40 mm was
64
present in 80 patients. Median echocadiographic follow-up was 6.7 (IQR 5.4-8.4) years,
65
and 100% complete. The main end-point of the study was postoperative TR of moderate or
66
greater degree.
67
Results: Thirty patients had new or persistent TR at one point during the follow-up. The
68
probability of postoperative TR at 7 years was 6.6% (95% CI 4.6 - 9.4) for all patients,
69
6.8% (95% CI 4.6 - 10.4) for TA <40mm, and 6.0% (95% CI 2.9 - 12.2) for TA ≥40 mm.
70
Preoperative TA diameter was not associated with odds of postoperative TR in either the
71
univariable or multivariable regression models. In these analyses preoperative TR was the
72
strongest predictor of postoperative TR.
73
Conclusion: TA≥40 mm did not predict the development of postoperative TR after MV
74
repair for degenerative diseases.
AC C
EP
TE D
M AN U
SC
RI PT
54
4
ACCEPTED MANUSCRIPT
Functional tricuspid regurgitation (TR) is due to leaflet tethering secondary to
76
right ventricular and annular dilatation.1 Right ventricular dilatation is often caused by
77
increased right ventricular pressure and/or volume overload, but it can also be caused by
78
right ventricular infarction or idiopathic cardiomyopathy. Functional TR is common in
79
patients with advanced left sided heart valve diseases. Current guidelines on the
80
management of heart valve diseases2-3 suggest that tricuspid valve annuloplasty (TVA)
81
may be beneficial in patients with trace or mild functional TR and a tricuspid annulus (TA)
82
diameter ≥40 mm by echocardiography or 70 mm by direct inspection of the tricuspid
83
valve (TV) during surgery for left sided valve lesions (Class 2a, which means that it should
84
be considered but the evidence is conflicting). The concept of performing TVA for TA
85
dilatation rather than the TR grade was introduced by Dreyfus and colleagues4 who
86
hypothesized that dilated TA was a permanent and even progressive lesion that if left
87
unattended at the time of mitral valve (MV) repair it may progress and cause severe TR
88
postoperatively. There is evidence that the TA dilates and changes its shape from elliptical
89
to spherical in patients with severe TR.1,5 What remains controversial is whether TVA
90
should be performed or not if the TR diameter measures 40 mm or more.
EP
TE D
M AN U
SC
RI PT
75
Degenerative diseases of the MV is the most common cause of mitral regurgitation
92
(MR) in North America but include a broad spectrum of lesions in the leaflets and annulus
93
of the atrioventricular valves. We hypothesize that a TA diameter ≥40 mm in patients with
94
MR due to degenerative diseases does not predict postoperative TR after MV repair for
95
degenerative diseases. This study examines the effect of the diameter of the TA in the
96
development of TR after MV repair for MR due to degenerative diseases.
97
Patients and methods
AC C
91
5
ACCEPTED MANUSCRIPT
98 99
This study was approved by the Review Ethics Board of University Health Network and individual patient consent was required. All patients (1,247) who had MV repair for MR due to degenerative diseases from January 1985 through December 2010 by
101
one surgeon (TED) have been followed prospectively. Patients operated since October
102
2005 had preoperative TA diameters measured in systole and diastole. This new entry was
103
prompted by the publication by Dreyfus and colleagues on the importance of the
104
preoperative TA size in predicting postoperative TR after MV repair.4 Of 335 patients
105
operated on from October 2005 through December 2010, 25 had concomitant TVA and
106
were excluded from this study. Patients were followed by the referring cardiologists and
107
seen by the surgeon at 2 to 3 months and contacted by his research personnel every 2 to 3
108
years thereafter. Echocardiography was performed preoperatively, during surgery, and
109
postoperatively at one week, and at every follow-up contact unless there was a reason to
110
image the heart more often. A total of 984 postoperative echocardiograms were available
111
for analysis: 303 (97%) patients had more than one study and 87% had more than 2
112
studies. Most postoperative studies were performed at certified echocardiography
113
laboratories and read by an experienced echocardiographer. MR and TR were recorded as
114
none (0), trace (1+), mild (2+), moderate (3+) and severe (4+), according to the guidelines
115
described by the American Society of Echocardiography.6 If the echocardiographer
116
reported the MR or TR as “mild to moderate” or “moderate to severe”, which occurred in
117
27 studies, the recorded images were reviewed by blinded echocardiographers from our
118
institution and entered into the database according to their interpretation as mild (6
119
patients), moderate (19 patients) or severe (2 patients). The diameters of the tricuspid and
120
mitral annuluses were measured preoperatively by transesophageal echocardiography at
AC C
EP
TE D
M AN U
SC
RI PT
100
6
ACCEPTED MANUSCRIPT
end of diastole in the 4-chamber view at 0º for the TA and at 0º and 90º for the mitral
122
annulus (MA). Adverse events were recorded according to guidelines set by cardiac
123
surgical societies.7 The cause of death was determined by hospital charts review, death
124
certificates, or information from the physician who was caring for the patient at that time.
125
Clinical and echocardiographic follow-up was 100% complete; the median (IQR) duration
126
of clinical follow-up was 7.1 (5.8-8.8) years and the median (IQR) duration of
127
echocardiographic follow-up was 6.7 (5.4-8.4) years. There were no substantial differences
128
between the number of patients reaching landmark durations of follow-up when comparing
129
clinical and imaging follow-up.
130
Statistical analysis
M AN U
SC
RI PT
121
Data is presented as median with interquartile range (IQR, 25th and 75th percentiles)
132
and frequencies as appropriate. Pearson correlation was used to determine the association
133
between preoperative TA and MA size along with linear regression models. Non-linearity
134
of the association between preoperative TA and MA size was assessed through various
135
mathematical transformations of MA size; the Akaike Information Criteria (AIC) was used
136
to compare the different models. Linear regression was used to determine the association
137
between preoperative TA, preoperative TR and degree of myxomatous degeneration
138
(separately, both modelled as categorical variables). Kaplan-Meier method was used to
139
estimate survival over time. Cumulative incidence of stroke and reoperation were
140
calculated with death as a competing risk. Given the low number of patients experiencing
141
any of those 3 outcomes, no attempts at identifying risk factors was performed. The
142
progression of moderate or severe TR over time was modelled using logistic regression
143
models adjusted for repeated measures through an autoregressive covariance structure
AC C
EP
TE D
131
7
ACCEPTED MANUSCRIPT
(mixed longitudinal regression models). The autoregressive covariance structure was chosen
145
specifically because this structure assumes homogenous variance over time and correlations
146
between the different observations on the same patient that decline exponentially over time. Both
147
assumptions are true in this type of data (repeated echocardiograms over time). Univariable risk
148
factor analysis was performed using this approach. Given the limited number of outcomes available
149
for analysis, we elected to perform multivariable risk factor analysis using a priori selection of
150
variables based on clinical relevance. Variables selected as potential risk factors were: sex, age at
151
the time of surgery, preoperative atrial fibrillation/concomitant Maze procedure, preoperative
152
hypertension, preoperative TA ≥40 mm, preoperative TR grade, and time since index procedure.
153
The association between TA size ≥40 mm and progression of moderate or severe TR over
154
time was assessed for various subgroup of patients to determine whether there was any
155
interaction between clinical characteristics and the effect of preoperative TA size on
156
postoperative TR.
SC
M AN U
TE D
157
RI PT
144
Missing data was rare (2.0%) and randomly distributed, in consequence mean imputation was used to handle the few missing data elements; there was no missing outcomes
159
data. All statistical analyses were performed using SAS v9.4 (SAS statistical software, Cary NC).
160
Results
161
Patient profile and operative data for the entire cohort and stratified by preoperative TA
162
size are shown in Tables 1 and 2 respectively.
163
Clinical outcomes - There were 2 operative and 11 late deaths: 2 due to stroke, 1
164
endocarditis and 10 non-cardiac deaths. There was one MV re-repair for recurrent MR 5
165
years after surgery and one MV re-repair for mitral stenosis caused by pannus 7 years after
AC C
EP
158
8
ACCEPTED MANUSCRIPT
surgery. Six patients suffered a stroke (2 died). Table 3 shows the freedom from morbid
167
events over time.
168
Tricuspid and mitral valves – There was a modest positive linear association between
169
preoperative TA and MA diameters as shown in Figure 1 and between TA diameter and
170
TR grade as shown in Figure 2, but no association between TA diameter and the degree of
171
degeneration of the MV as seen in Figure 3. Eighty patients (26%) had TA ≥40 mm before
172
surgery. Thirty patients had new or persistent TR ≥3+ at one point during the follow-up.
173
Figure 4 shows the probability of postoperative TR≥3+ during the follow-up in all patients,
174
in patients with TA≥40 mm, and in patients with TA <40 mm separately. Table 4 shows
175
the results of univariable model for variables associated with postoperative TR≥3+ and
176
Table 5 shows the results of multivariable model. Preoperative TA diameter was not
177
associated with odds of postoperative TR in either the univariable or multivariable
178
regression models. This remained true regardless of whether TA size was modelled as a
179
continuous variable, an ordinal variable or using the threshold of ≥40 mm. There was a
180
significant association between TA size indexed to body surface area and odds of
181
postoperative TR in univariable analysis (Table 4), however, this association did not
182
remain significant in the multivariable regression model (Table 5). In subgroup analysis
183
(Table 6), preoperative TA diameter ≥40 mm was not associated with increased risk of
184
postoperative TR in any subgroup of patient except for patients with preoperative ejection
185
fraction <60%.
186 187
AC C
EP
TE D
M AN U
SC
RI PT
166
Twelve patients developed MR ≥3 and one patient developed mitral stenosis due to pannus during the follow-up.
9
ACCEPTED MANUSCRIPT
188
189
Discussion The concept of performing TVA based on the diameter of the TA rather than the severity of TR was introduced by Dreyfus and colleagues4 in 2005. These investigators
191
systematically explored the TV at the time of MV repair for MR due to various pathologies
192
in 311 patients and performed TVA whenever the distance from the antero-septal
193
commissure to the antero-posterior commissure was ≥70 mm.4 They showed that this
194
approach greatly reduced the risk of developing TR after MV repair.4 We are not aware of
195
any study that correlated this surgical measurement in the arrested heart with preoperative
196
TA diameter, but 40 mm appeared in the guidelines.2-3 Although we do not dispute that a
197
dilated TA is often associated with functional TR, the value 40 mm as a cut off to
198
recommend TVA without TR at the time of MV repair for degenerative MR needs further
199
evaluation. The TV leaflets like the MV leaflets vary in size in patients with degenerative
200
diseases. Patients with fibroelastic deficiency frequently have small MV leaflets, so small
201
that at one time we were reluctant to use an annuloplasty ring or band from fear of creating
202
mitral stenosis.8 The TV leaflets are also small in these patients and a TA of 40 mm is
203
probably too large and TVA is probably necessary to prevent late TR. On the other hand,
204
patients with myxomatous degeneration of the MV frequently have large leaflets with large
205
MA and the TV leaflets are also large with consequent larger annuluses. This type of
206
degenerative disease is present in most of our patients and a TA of 40 mm is probably
207
normal and seldom are associated with severe TR (at least in our experience), and when
208
they are, it is often because of leaflet prolapse rather than annular dilatation. In this study,
209
the diameter of the TA correlated modestly with the size of the MA (Figure 1), as well as
210
with the severity of preoperative TR (Figure 2) but not with the degree of myxomatous
AC C
EP
TE D
M AN U
SC
RI PT
190
10
ACCEPTED MANUSCRIPT
degeneration, which surprised us because we expected that patients with more advanced
212
degrees of myxomatous degeneration would have large TV leaflets and consequently
213
larger TA.
214
RI PT
211
TA ≥40 mm or TA ≥21 mm/m2 did not predict the development of postoperative TR. Preoperative TA diameter was not associated with postoperative TR in either the
216
univariable or multivariable regression models. This remained true regardless of whether
217
TA diameter was modelled as a continuous variable, an ordinal variable or using the
218
threshold of ≥40 mm. There was a significant association between TA size indexed to body
219
surface area and odds of postoperative TR in univariable analysis, however, this
220
association did not remain significant in the multivariable regression model. Preoperative
221
TA diameter ≥40 mm was not associated with increased risk of postoperative TR in any
222
subgroup of patient except for patients with preoperative ejection fraction <60%. In a
223
recent publication, we examined a much larger sample of patients and found that impaired
224
left ventricular function was associated with the development of postoperative TR.9 The
225
most powerful predictor of postoperative TR was the presence of untreated moderate TR
226
before surgery.
M AN U
TE D
EP
We are not the first ones to challenge the concept that TA≥40 mm predicts
AC C
227
SC
215
228
postoperative functional TR after MV repair for degenerative diseases. Sordelli and
229
colleagues10 from Milan, Italy, prospectively evaluated the predictive value of TA diameter
230
assessed by 3D transesophageal echocardiography (3D-TEE) in the development of TR in
231
706 patients who had isolated MV repair for MR due to degenerative diseases (77%
232
myxomatous – with its broad spectrum of lesions – and 23% fibroelastic deficiency).
233
Those investigators performed a detailed analysis of the diastolic and systolic TA 11
ACCEPTED MANUSCRIPT
diameters (antero-posterior and septo-lateral) and the outcomes of surgery after a mean
235
follow-up of 2 years (range 0.5 to 5 years). TR decreased by one grade in 32% of patients,
236
remained unchanged in 62% and increased by one grade in 5.5% but only 3 patients
237
developed moderate or severe TR. The authors concluded that “newly developed
238
significant TR is a rare event after successful repair of degenerative MR” and that
239
“analysis of TA does not predict early to midterm subsequent TR progression”. The MV
240
pathology (myxomatous vs fibroelastic deficiency) was not entered into the multivariable
241
analysis models in that study. It is noteworthy mentioning that the sizes of the TA and MA
242
in our study were very similar to those reported by Sordelli and colleagues.10
M AN U
SC
RI PT
234
A recent report based on the STS database on TVA at the time of MV surgery
244
indicated that concomitant TVA was performed in 14.3% of all MV operations but in
245
centers that performed more than 10 MV operations per year it ranged from 75% in 2
246
centers to less than 10% in more than 300 centers.11 In addition, TVA was performed in
247
only 3.5% when the grade of TR was mild or less, 30.6% when it was moderate and 75.6%
248
when severe.11 However, as documented in this study and a previous one9 preoperative
249
moderate TR is associated the postoperative TR of moderate or severe degree after MV
250
repair for degenerative diseases and we now bow believe that TVA is appropriate in these
251
patients. In addition, TVA should be considered in older patients with longstanding atrial
252
fibrillation, dilated atria, impaired left ventricular systolic function probably should have
253
TVA, particularly if they are women even in absence of moderate or severe TR.9, 12
254
Limitations of study – This is a retrospective study and the TA and MA diameters were
255
obtained intraoperatively and the values may have been affected by the anesthetic, assisted
256
ventilation, and loading conditions which were not controlled at the time of the
AC C
EP
TE D
243
12
ACCEPTED MANUSCRIPT
measurements. The size and area of the TV leaflets were not measured to correlate with the
258
diameter of TA and development of TR. In addition, TR after MV repair for degenerative
259
diseases is uncommon and the sample size of this study was too small and the duration of
260
follow-up too short for more detailed analysis of the variables associated with
261
postoperative TR. Finally, given the limited sample size and the close to significant
262
association between preoperative TA diameter as a continuous variable and increase odds
263
of postoperative TR over time we cannot exclude that the lack of association is in fact a
264
type II error resulting from an underpowered comparison.
265
Conclusions – TA≥40 mm was not associated with the development of TR after MV repair
266
for degenerative diseases, likely because of variable sizes of the TV leaflets in patients
267
with degenerative diseases. Preoperative moderate TR was the most powerful predictor of
268
postoperative TR by multivariable analysis.
271
272
273
SC
M AN U
TE D EP
270
AC C
269
RI PT
257
274
275
276 13
ACCEPTED MANUSCRIPT
277
278
References 1. Sagie A, Schwammenthal E, Padial LR, Vazquez de Prada JA, Weyman AE, Levine RA. Determinants of functional tricuspid regurgitation in incomplete tricuspid valve closure:
280
Doppler color flow study of 109 patients. J Am Coll Cardiol. 1994;24:446-53.
281
RI PT
279
2. Alec Vahanian A, Alfieri O, Andreotti F, Antunes MJ, Barón-Esquivia G, Baumgartner H., et al. Guidelines on the management of valvular heart disease (version 2012) - Eur J
283
Cardio-Thorac Surg 42 (2012) S1–S44.
3. Nishimura RA, Otto CM, Bonow RO, Carabello BA, Erwin III JP, Guyton RA, et al. 2014
M AN U
284
SC
282
285
AHA/ACC Guideline for the management of patients with valvular heart disease:
286
Executive Summary. J Am Coll Cardiol 2014;63:2438-88.
287
4. Dreyfus GD, Corbi PJ, Chan KMJ, and Bahrami T. Secondary Tricuspid Regurgitation or Dilatation: Which Should Be the Criteria for Surgical Repair? Ann Thorac Surg
289
2005;79:127-32.
290
TE D
288
5. Nemoto N, Lesser JR, Pedersen WR, Sorajja P, Spinner E, Garberich RF et al. Pathogenic structural heart changes in early tricuspid regurgitation. J Thorac Cardiovasc Surg
292
2015;150:323-30.
6. Zoghbi WA, Enriquez-Sarano M, Foster E, Grayburn PA, Kraft CD, Levine RA, et al.
AC C
293
EP
291
294
American Society of Echocardiography. Recommendations for evaluation of the severity
295
of native valvular regurgitation with two-dimensional and Doppler echocardiography. J
296
Am Soc Echocardiogr 2003;16:777-802.
297
7. Akins CW, Miller DC, Turina MI, Kouchoukos NT, Blackstone EH, Grunkemeier GL, et
298
al. Guidelines for reporting mortality and morbidity after cardiac valve interventions. J
299
Thorac Cardiovasc Surg 2008;135:732-738. 14
ACCEPTED MANUSCRIPT
301 302 303 304
8. David TE, Armstrong S, McCrindle BW, Manhiolt C. Late outcomes of mitral valve repair for mitral regurgitation due to degenerative disease. Circulation 2013;127:1485-92. 9. David TE, David CM, Fan CPS, Manhiolt C. Tricuspid regurgitation is uncommon after mitral valve repair for degenerative diseases. J Thorac Cardiovasc Surg 2017;154:110-22.
RI PT
300
10. Sordelli C, Lancellotti P, Carlomagno G, Di Giannuario G, Alati E, De Bonis M, et al. Tricuspid Annular Size and Regurgitation Progression After Surgical Repair for
306
Degenerative Mitral Regurgitation. Am J Cardiol 2016;118:424-31
307
SC
305
11. Badhwar V, Rankin S, He M, Jacobs JP, Furnary AP, Fazzalari FL et al. Performing concomitant tricuspid valve repair at the time of mitral valve operations is not associated
309
with increased operative mortality. Ann Thorac Surg 2017;103:587-593.
310
M AN U
308
12. Ro SK, Kim JB, Jung SH, Choo SJ, Chung CH, Lee JW. Mild-to-moderate functional tricuspid regurgitation in patients undergoing mitral valve surgery. J Thorac Cardiovasc
312
Surg 2013;146:1092-7.
315 316
317
318
EP
314
AC C
313
TE D
311
319
320 15
ACCEPTED MANUSCRIPT
Legends
322
Figure 1: Correlation of preoperative tricuspid annulus size vs. preoperative mitral
323
annulus size
324
Figure 2: Tricuspid annulus size vs. preoperative tricuspid regurgitation grade
325
Figure 3: Tricuspid annulus size vs. degree of degeneration of the mitral valve
326
Figure 4 - Estimated punctual probability of postoperative tricuspid regurgitation of
327
moderate degree or greater over time (figure) and at specific follow-up landmarks
328
(embedded table). Data provided for in all patients (upper panel) and in patients with TA
329
<40 mm TA≥40 mm
M AN U
SC
RI PT
321
333
334
335
336
EP
332
AC C
331
TE D
330
337
338 16
ACCEPTED MANUSCRIPT
339 340 341 342
Table 1: Patients profile stratified by preoperative tricuspid annulus diameter
343
Median age in years [IQR]
344
Male sex
345
Body surface area in m2 [IQR]
1.91 [1.75-2.08]
1.90 [1.72-2.06] 2.00 [1.85-2.11]
<0.001
346
Body mass index in kg/m2 [IQR]
25.9 [23.6-28.4]
25.8 [23.2-28.4] 26.3 [24.2-28.6]
0.42
347
Pre-operative heart rhythm
Entire cohort) (N=312) 58 [50-67]
Atrial fibrillation or flutter
44 (14.1)
349
History of atrial fibrillation
350
Complete heart block/pacemaker
59 [49-67]
57 [50-66]
151 (65.1)
P value
0.89
70 (87.5)
<0.001
25 (10.8)
19 (23.8)
0.008
10 (3.2)
7 (3.0)
3 (3.8)
0.75
1 (0.3)
0 (0.0)
1 (1.3)
0.26
M AN U
348
TA ≥40mm (N=80)
SC
221 (70.8)
TA <40mm (N=232)
RI PT
Variable
Previous heart operations
352
Coronary artery bypass
2 (0.6)
2 (0.9)
0 (0.0)
1.00
353
Percutaneous coronary interventions 2 (0.6)
2 (0.9)
0 (0.0)
1.00
354
Atrial septal defect closure
2 (0.9)
0 (0.0)
1.00
355
NYHA functional classification
357
Class II
358
Class III
359
Class IV
360
2 (0.6)
0.24
93 (29.8)
72 (31.0)
21 (26.3)
124 (39.7)
96 (41.4)
28 (35.0)
91 (29.2)
62 (26.7)
29 (36.3)
4 (1.3)
2 (0.9)
2 (2.5)
EP
Class I
AC C
356
TE D
351
Left ventricular ejection fraction
0.08
361
≥ 60%
258 (82.7)
198 (85.3)
60 (75.0)
362
40-59%
46 (14.7)
28 (12.1)
18 (22.5
363
20-39%
8 (2.6)
6 (2.6)
2 (2.5)
108 (34.6)
76 (32.8)
32 (40.0)
364 365
Symptomatology Congestive heart failure
17
028
ACCEPTED MANUSCRIPT
Angina pectoris
367
Syncopal episodes
13 (4.2)
13 (5.6)
0 (0.0)
0.03
4 (1.3)
4 (1.7)
0 (0.0)
0.58
23 (28.8)
368
Associated conditions
369
Hyperlipidemia
111 (35.6)
88 (37.9)
370
Family history of CAD
135 (43.3)
106 (45.7)
371
Diabetes mellitus
15 (4.8)
12 (5.2)
372
Hypertension
115 (36.9)
87 (37.5)
373
Smoking history
124 (39.7)
90 (38.8)
374
Severe COPD
5 (1.6)
375
Previous stroke or TIA
376 377
0.18
RI PT
366
0.15
3 (3.8)
0.77
28 (35.0)
0.79
34 (42.5)
0.60
4 (1.7)
1 (1.3)
1.00
9 (2.9)
9 (3.9)
0 (0.0)
0.12
Infective endocarditis
14 (4.5)
12 (5.2)
2 (2.5)
0.53
Coronary artery disease
43 (13.8)
31 (13.4)
12 (15.0)
0.71
M AN U
SC
29 (36.3)
378 379 Numbers in parentheses are percentages.
381 382
Abbreviatons: IQR = interquartile range; NYHA = New York Heart Association; CAD = coronary artery disease; COPD = chronic obstructive pulmonary disease; TIA = transient ischemic episode
385 386 387 388
EP
384
AC C
383
TE D
380
389 390 391
18
ACCEPTED MANUSCRIPT
392
Table 2. Operative data stratified by preoperative tricuspid annulus diameter
393 394 395
Variable
Entire cohort) (N=312)
TA <40mm (N=232)
Mitral valve prolapse 11 (4.7)
397
Anterior leaflet
21 (6.7)
14 (6.0)
398
Posterior leaflet
112 (35.9)
84 (36.2)
399
Both leaflets
167 (53.5)
123 (53.0)
Echocardiographic measurements: Mitral annulus at 0° in mm [IQR]
40 [37-45]
402
Mitral annulus at 90° in mm [IQR]
40 [37-45]
403
Tricuspid annulus at 0° in mm [IQR]
35 [32-40]
Tricuspid regurgitation grades:
40 [36-43]
7 (8.8)
28 (35.0)
44 (55.0)
43 [39-47]
<0.001
40 [36-44]
42 [39-47]
0.02
34 [32-36]
41 [40-42]
<0.001
M AN U
401
1 (1.3)
RI PT
12 (3.9)
SC
None (annular dilatation)
404
0.06
None or trace (0 – 1+)
150 (48.1)
120 (51.8)
30 (37.5)
406
Mild (2+)
144 (46.2)
101 (43.5)
43 (53.8)
407
Moderate (3+)
18 (5.8)
11 (4.7)
7 (8.8)
TE D
405
Degree of mitral valve myxomatous degeneration* Mild
410
Moderate
411
Advanced
AC C
409
EP
408
P value 0.46
396
400
TA ≥40mm (N=80)
412
Dystrophic calcification of the annulus
413
Operation performed:
0.94
105 (33.7)
79 (34.1)
26 (32.5)
135 (43.3)
99 (42.7)
36 (45.0)
72 (23.1)
54 (23.3)
19 (22.5)
12 (3.9)
10 (4.3)
2 (2.5)
0.74
414
MV repair with Simplici-T band
312 (100)
232 (100)
80 (100)
1.00
415
Chordal replacement
252 (80.8)
184 (79.3)
68 (85.0)
0.32
416
Reconstruction of the mitral annulus
12 (3.9)
10 (4.3)
2 (2.5)
0.74
417
19
ACCEPTED MANUSCRIPT
Maze for atrial fibrillation
54 (17.3)
32 (13.8)
22 (27.5)
0.01
419
Repair of atrial septal defect
15 (4.8)
9 (3.9)
6 (7.5)
0.23
420
Replacement of the ascending aorta
7 (2.2)
6 (2.6)
1 (1.3)
0.68
421
Coronary artery bypass
43 (13.8)
31 (13.4)
12 (15.0)
0.71
422
Resection of atrial myxoma
1 (0.3)
1 (0.4)
423
Septal myectomy
2 (0.6)
1 (0.4)
66 [47-81]
63 [46-100]
425
.
426
Cardiopulmonary bypass time
427
Median [IQR], minutes
Median [IQR], minutes
82 [64-101]
428 Numbers in parentheses are percentages.
430
* Degree of myxomatous degeneration from reference #8
431
Abbreviatons: IQR = interquartile range
434 435 436 437 438 439
EP
433
AC C
432
TE D
429
440 441 442 20
0 (0.0)
1.00
1 (1.3)
0.45
SC
Aortic clamping time
M AN U
424
RI PT
418
81 [62-100]
72 [53-84]
0.06
92 [71-106]
0.04
ACCEPTED MANUSCRIPT
443 444 445
Table 3: Clinical Outcomes - Freedom from mortality from Kaplan-Meier estimates, cumulative incidence of reoperation and stroke over time with death as a competing risk and number of patients with data available at different follow-up landmarks (95% confidence interval, CI)
446
Outcomes
1 year (95% CI)
447
Survival
99.4 (97.5-99.8)
98.4 (96.1-99.3)
448
Reoperation
0.0
449
Stroke
450 451
90.8 (74.2-96.9)
0.7 (0.2-2.6)
0.7 (0.2-2.6)
0.7 (0.2-2.6)
0.3 (0.1-2.3)
0.5 (0.2-2.5)
1.3 (0.5-3.5)
2.9 (1.1-7.3)
N (clinical)
310
297
277
N (imaging)
301
289
255
M AN U
452 453 454 455 456
461 462 463 464 465
EP
460
AC C
459
TE D
457 458
RI PT
97.7 (95.2-98.9)
SC
3 years (95% CI) 5 years (95% CI) 10 years (95% CI)
466 467 468
21
16
10
ACCEPTED MANUSCRIPT
Table 4: Univariable risk factors analysis for postoperative TR ≥3+ over time (odds ratio, OR with 95% confidence interval, CI)
471
Variable
OR
lower CI
upper CI
P-value
472
Age (5-year increment)
1.61
1.34
1.93
<0.001
473
Time since operation (years)
1.05
0.96
1.14
0.27
474
Preoperative MA size (mm)
0.99
0.92
1.06
0.72
475
Preoperative MV area in 0 degrees
0.99
0.93
1.06
0.85
476
Preoperative MV area in 90 degrees 0.96
0.89
1.03
0.22
477
Sex: female
4.08
1.84
9.05
<0.001
478
Atrial fibrillation/Maze procedure
4.63
2.10
10.24
<0.001
479
Hypertension
2.90
1.29
6.55
0.01
480
Preoperative TA size (mm)
1.06
0.99
1.15
0.10
0.88
6.17
0.09
1.74
0.57
5.31
0.33
1.68
0.20
14.02
0.63
3.71
0.44
31.18
0.23
M AN U
SC
RI PT
469 470
<35 mm
reference
482
35-39 mm
2.33
483
40-42 mm
484
42-45 mm
485
>45mm
TE D
481
Preoperative TA size ≥ 40 mm
1.13
0.48
2.68
0.77
487
TA size / BSA > 21 mm/m2
2.416
1.279
4.565
0.007
488
Preoperative TR ≥3+
13.6
5.71
32.59
<0.001
489
Preoperative degeneration of MV (ref.: mild)
490
Moderate
0.86
0.36
2.06
0.74
491
Severe
0.39
0.14
1.07
0.07
492
Preoperative LV grade (per grade)
AC C
EP
486
1.03
0.49
22
2.17
0.94
ACCEPTED MANUSCRIPT
493 494 495
Table 5: Multivariable risk factors analysis for postoperative TR ≥3+ over time for preoperative TA size ≥40mm and preoperative TA size indexed to body surface area (odds ratio, OR with 95% confidence interval, CI)
496
Variable
497
Preoperative TA size ≥40 mm
498
Time since operation (per year)
1.094
0.986
1.213
0.09
499
Age (5-year increment)
1.312
1.054
1.633
0.02
500
Sex (female)
4.244
1.748
10.307
0.001
501
Atrial fibrillation/Maze procedure
1.744
0.666
4.567
0.26
502
Hypertension
2.485
0.958
503
Preoperative TR ≥3+
7.629
2.105
504
Preoperative TA size ≥ 40 mm
1.405
0.492
505
P-value
RI PT
upper CL
SC
lower CL
6.449
0.06
27.658
0.002
4.014
0.53
M AN U
OR
Preoperative TA indexed to body surface area
507
Time since operation (per year)
1.094
0.986
1.213
0.09
508
Age (5-year increment)
1.294
1.050
1.594
0.02
509
Sex (female)
3.735
1.445
9.652
0.007
510
Atrial fibrillation/Maze procedure
1.849
0.680
5.024
0.23
511
Hypertension
2.513
0.965
6.546
0.06
512
Preoperative TR ≥3+
7.537
2.141
26.528
0.002
513
Preoperative TA indexed to BSA
1.024
0.905
1.160
0.70
516 517 518 519
EP
515
AC C
514
TE D
506
520 521 522
23
ACCEPTED MANUSCRIPT
523 524
Table 6: Association between preoperative tricuspid annulus size ≥40mm and postoperative TR ≥3+ over time for various sub-groups of patients (odds ratio, OR with 95% confidence interval, CI)
525
Patient subgroup
526
Age (years)
Lower CL
Upper CL
<50 years
3.376
0.214
53.372
528
50-59 years
0.908
0.750
10.930
529
60-69 years
0.626
0.130
3.011
530
≥70 years
2.013
0.605
6.701
531
Sex
0.39
0.94
0.56
0.25
SC
527
P-value
RI PT
OR
Male
1.191
0.366
3.876
0.77
533
Female
3.210
0.883
11.666
0.08
534
Atrial fibrillation/Maze procedure*
535
Yes
1.472
536
No
0.474
537
Body surface area (m2)
M AN U
532
0.474
4.573
0.50
0.102
2.207
0.34
0.399
9.599
0.40
≤1.75
1.993
539
1.76-1.90
1.052
0.238
4.663
0.95
540
1.91-2.05
1.246
0.114
13.591
0.86
541
>2.05
2.158
0.403
11.547
0.37
1.855
0.310
11.115
0.50
1.518
0.449
5.135
0.50
0.229
0.045
1.167
0.08
Preoperative grade of TR 1-2
544
2
545
≥3+
546
AC C
543
EP
542
TE D
538
Preoperative degeneration of MV
547
Mild
1.540
0.406
5.847
0.53
548
Moderate
0.777
0.200
3.017
0.72
549
Severe
1.642
0.310
8.695
0.56
550
Ejection fraction (%)
551
≥60%
0.672
0.242
1.860
0.44
552
<60%
7.112
1.205
41.980
0.03
24
ACCEPTED MANUSCRIPT
553
Preoperative MA size (mm) ≤37 mm
1.713
0.410
7.151
0.46
555
38-40 mm
2.536
0.257
25.013
0.43
556
41-45 mm
0.264
0.029
2.390
0.24
557
≥46 mm
1.626
0.262
10.091
558
0.60
*Given the small number of patients, the 41 patients with AF and Maze procedure, 13 patients with history of AF
EP
TE D
M AN U
SC
and Maze procedure and the 3 patients with AF but no Maze procedures were combined in a single category.
AC C
559 560
RI PT
554
25
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT