Novel Trigone-Based Sizing Method for Mitral Ring Annuloplasty

Journal Pre-proof Novel trigone-based sizing method for mitral ring annuloplasty Hiroshi Okamoto, MD, Yasuyuki Fujimoto, MD, Chikao Teramoto, MD PII:

S0003-4975(19)31476-6

DOI:

https://doi.org/10.1016/j.athoracsur.2019.08.085

Reference:

ATS 33108

To appear in:

The Annals of Thoracic Surgery

Received Date: 23 December 2018 Revised Date:

28 July 2019

Accepted Date: 19 August 2019

Please cite this article as: Okamoto H, Fujimoto Y, Teramoto C, Novel trigone-based sizing method for mitral ring annuloplasty, The Annals of Thoracic Surgery (2019), doi: https://doi.org/10.1016/ j.athoracsur.2019.08.085. This is a PDF file of an article that has undergone enhancements after acceptance, such as the addition of a cover page and metadata, and formatting for readability, but it is not yet the definitive version of record. This version will undergo additional copyediting, typesetting and review before it is published in its final form, but we are providing this version to give early visibility of the article. 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. © 2019 by The Society of Thoracic Surgeons

Hiroshi Okamoto

Novel trigone-based sizing method for mitral ring annuloplasty Running head; Trigone-based ring sizing

Hiroshi Okamoto, MD, Yasuyuki Fujimoto, MD, Chikao Teramoto, MD

Department of Cardiovascular Surgery, Yokkaichi Municipal Hospital, Yokkaichi, Japan

Address for correspondence and reprint requests: Hiroshi Okamoto, MD, Department of Cardiovascular Surgery, Yokkaichi Municipal Hospital, 2-22-37 Shibata, Yokkaichi, Japan. E-mail: [email protected]

-1–

Hiroshi Okamoto

Abstract Background: We devised a novel trigone-based sizing method, setting the trigones at 1/4 of annular circumference, and used it for mitral annuloplasty in patients with mitral regurgitation (MR). Methods: Between 1999 and 2017, 436 patients with degenerative (N=192), non-ischemic functional (N=124), or ischemic (N=120) MR underwent mitral valvuloplasty at our institution using an incomplete ring. The inter-trigonal distance and pre/post-repair annular diameter were measured. Then the diameters predicted from body surface area and the inter-trigonal distance, and the ratios of these diameters to observed data, were computed. We investigated the influence of these measurements on MR recurrence, trans-mitral pressure gradient, and systolic anterior motion. Results: Initial repair was successful in 433 patients (99%), but three patients with systolic anterior motion and MR required conversion to valve replacement. After 1, 5, and 10 years (mean follow-up: 6.3 years), the rate of freedom from grade ≥2 recurrent MR was respectively 96%, 92%, and 86% in the degenerative group, 99%, 97%, and 90% in the non-ischemic functional group, and 95%, 90%, and 79% in the ischemic

-2–

Hiroshi Okamoto

group (p=0.052). The observed/body surface area predicted diameter ratio was negatively correlated with the mean trans-mitral pressure gradient (mmHg); 12.3 – 8.2 ∗ (ratio) (R=-0.37, p<0.001), despite a smaller ratio (<0.9) not being associated with less recurrence of MR. In the degenerative group, 7/71 patients (10%) with an observed/inter-trigonal distance predicted diameter ratio <0.9 developed systolic anterior motion (p<0.001). Conclusions: Our trigone-based sizing method achieved satisfactory control of MR, while avoiding functional mitral stenosis and systolic anterior motion.

-3–

Hiroshi Okamoto

Ring annuloplasty is usually required when mitral valvuloplasty (MVP) is performed for chronic MR because annular dilation occurs in degenerative or functional MR [1-3]. Choosing the optimum ring size is critical to control MR, while avoiding functional mitral stenosis (MS) [3] or systolic anterior motion (SAM) [4]. Current commissureand trigone-based sizing strategies of the manufacturers recommend setting commissures or trigones at 1/3 of annular circumference [4-7], but these strategies contradict each other, because the trigones and commissures are located at different distances. Another proposed method involves matching the sizer to an anterior mitral leaflet (AML) [1], but it has the problem that there is no clear definition of AML size [6]. When we began using the Cosgrove band (Edwards Lifesciences, Irvine, CA), we were surprised to find that the Carpentier ring sizer was recommended for sizing [7]. We subsequently studied cadaver hearts without mitral valve disease to determine the normal locations of the commissures and trigones in the annulus and how AML size should be represented. We obtained the following findings: 1) the trigones are positioned at 1/4 of the undilated annulus and the proportion of inter-trigonal region decreases with annulus enlargement, 2) the commissures are located at 1/3 of annular

-4–

Hiroshi Okamoto

circumference irrespective of enlargement, and 3) AML size can be represented by AML-Quad, which is a quadrilateral formed by lines connecting the trigones and the attachments of the two main papillary muscles [8]. These findings suggested that the commissures may be inappropriate for ring sizing and that setting the trigones at 1/3 of annular circumference may lead to over-reduction. We then devised an original sizer that sets the trigones at 1/4 of annular circumference and applied it to MVP using an incomplete ring (partial band). In the present study, we verified the validity of this sizing method by examining the relationship between the post-repair annular diameter (Dm) and postoperative recurrence of MR, functional MS, and SAM.

Patients and Methods Patient population Between January 1999 and November 2017, 436 consecutive patients underwent MVP for MR at our institution using an incomplete ring (partial band) (421 Cosgrove and 15 Tailor bands (Abbott, St. Paul, MN)). The patients were divided into three groups based on the etiology of MR, which were the degenerative MR (DMR) group (N=192), the

-5–

Hiroshi Okamoto

non-ischemic functional MR (NI-FMR) group (N=124), and the ischemic MR (IMR) group (N=120). In the NI-FMR group, MR was mainly secondary to aortic valve disease along with various other causes. All IMR patients had old myocardial infarction. After discharge, patients visited our outpatient clinic for follow-up once a year. The follow-up rate for assessment of survival was 99% and the mean follow-up period was 6.3 ± 4.9 (0.1-18.8) years. Freedom from recurrence of grade ≥2 MR were evaluated up to the end of 2017. This study was approved by our Ethics Committee, and the committee waived the need for patient consent because our study only involved retrospective analysis of anonymized data.

Surgical procedures All operations were performed in a routine fashion via median sternotomy by a single surgeon (HO). Up to 2009, posterior leaflet prolapse was corrected in DMR patients by resection and suture [1], while AML prolapse was managed with artificial chordae (CV-3 or 4 Gore-Tex sutures, W.L. Gore & Associates, Flagstaff, AZ) (N= 96). Since 2010, all prolapsed leaflets have been repaired with artificial chordae (N= 96) [5],

-6–

Hiroshi Okamoto

because it was found that resection and suture was inadequate treatment for extensive prolapse and there was a risk of disruption requiring reoperation. In the NI-FMR and IMR groups, simple annuloplasty was employed for 179 patients (73%) and 65 patients (27%) received an additional relocation technique [9] from 2010 to 2016. If severe tethering (tenting height >10 mm) was noted, mitral valve replacement (MVR) was done [10]. The concomitant cardiac procedures are summarized in Table 1.

Sizing & Implantation method After unfurling the AML, ridges were clearly visualized on its surface running between the trigones and the attachments of the two main papillary muscles. These ridges formed a quadrilateral that we designated as the AML-Quad (yellow shape in Figure 1A). The trigones (and commissures) were marked with crystal violet and the straight part of the sizer was fitted to the inter-trigonal region (Lt: yellow line with arrows, Figure 1B). When an incomplete ring is implanted, it should be stitched to the trigones at both ends (Figure 1C) (video). The curved part of the sizer corresponded to the internal border length of appropriate ring for the patient (Figure 2A-C). The functional mitral annulus

-7–

Hiroshi Okamoto

(internal border of ring) lies inside the native annulus (ring centerline) by the ring width. (Figure 2B-C). After implantation of the ring, the functional annulus was reduced in size until it just circumscribed the AML-Quad (Figure 2C).

Mitral valve dimensions and clinical implications The observed Dm was measured with graduated intubation tubes before and after repair [8]. The inter-trigonal distance (length) was measured on the screen (observed Lt, Figure 1A) by 2 independent observers to reduce potential bias, and the mean value was calculated (R=0.94, p<0.001). The post-repair diameter was calculated from the band length and width (calculated Dm, Figure 2C). The diameter and inter-trigonal distance predicted based on ideal body surface area (BSA) (BSA-predicted Dm and Lt) were calculated according to the method of Rowlatt [11-13]. The diameter predicted from observed Lt ((Lt-predicted Dm); post-repair target diameter) was calculated based on the diagram shown in Figure 2C. The ratio of the observed diameter to the calculated diameter (Dm (observed/calculated) ratio) was determined, as well as the ratio to the BSA-predicted diameter (Dm (observed/BSA-predicted) ratio) and the Lt-predicted

-8–

Hiroshi Okamoto

diameter (Dm (observed/Lt-predicted) ratio). The formulae used for calculation are summarized at the end of the text. We evaluated the clinical implications of the following post-repair Dm ratios: (1) Dm (observed/calculated) ratio for size reproducibility (2) Dm (observed/Lt-predicted) ratio for occurrence of SAM (3) Dm (observed/BSA-predicted) ratio for recurrence of MR and functional MS (trans-mitral pressure gradient (TMPG)).

Echocardiography Using two-dimensional transthoracic echocardiography (TTE) or transesophageal echocardiography (TEE), MR was classified into 5 grades [0 (none or trivial), 1 (mild), 2 (mild to moderate), 3 (moderate), or 4 (severe)] according to American Society of Echocardiography guidelines [14]. After weaning from cardiopulmonary bypass, residual MR was evaluated by TEE. After surgery, TTE was performed before discharge, at 6 months after discharge, and then once a year. Various echocardiographic parameters are summarized in Table 2. At 1 year after surgery, 373 patients (82.5%) underwent

-9–

Hiroshi Okamoto

echocardiography, but the number declined to 213 (81.9%) at 5 years and to 118 (58.1%) at 10 years.

Statistical Analysis Statistical analysis of continuous variables was performed with parametric tests, including the t-test, one-way analysis of variance (ANOVA), and analysis of covariance (ANCOVA), or with non-parametric tests (Mann-Whitney U test or Kruskal-Wallis test). The chi-square test (or Fisher’s exact test for n<5) was employed for categorical variables. Correlations were calculated by Pearson’s or Spearman’s method. Cumulative freedom from late recurrence of MR were estimated by the Kaplan-Meier method and were compared between groups using the log-rank test. Statistical significance was defined at p<0.05. All analyses were performed with EZR on R commander v1.36 [15].

Results Patient characteristics and surgical data The NI-FMR and IMR groups had a higher prevalence of patients who were elderly, had

- 10 –

Hiroshi Okamoto

diabetes, and were in preoperative New York Heart Association (NYHA) functional class 3 or 4 (Table 1). We used Cosgrove bands ranging from size 26 to 36 and Tailor bands ranging from size 29 to 33. Up to 2010, the band size was selected according to the centerline diameter (mainly Cosgrove 28/30), resulting in accidental under-sizing with a higher mean TMPG (e. g., when the target post-repair diameter is 2.5 cm, a Cosgrove band width of 0.4 cm results in approximately 16% diameter (29% area) reduction). After identifying this problem of diameter loss associated with band width (Figure 2C), we subsequently selected the bands according to the internal border diameter (Cosgrove 32/34 and Tailor 29).

Echocardiography findings Preoperative mean LVEF was 62.3% in the DMR group, 48.4% in the NI-FMR group, and 34.7% in the IMR group, while grade 4 MR was present in 82%, 15%, and 21%, respectively (p<0.001). After surgery, the severity of MR decreased significantly in all groups, as did LVDd, LVDs, and systolic pulmonary arterial pressure. SAM occurred in

- 11 –

Hiroshi Okamoto

7 patients from the DMR group; 3 patients with Barlow disease required conversion to MVR at the second pump run, while the others responded to conservative treatment. Echocardiography data are summarized in Table 2.

Outcomes In all groups, the postoperative NYHA functional class of hospital survivors showed significant improvement (Table 1). Freedom from grade ≥2 recurrent MR at 1, 5, and 10 years after surgery was respectively achieved in 96%, 92%, and 86% of the DMR group, 99%, 97%, and 90% of the NI-FMR, and 95%, 90%, and 79% of the IMR group (p=0.052) (Figure 3A). Use of relocation techniques was discontinued from the end of 2016, because there was no preventive effect on grade ≥2 MR recurrence (the 5-year freedom rate was 79% vs. 90% for annuloplasty alone, p=0.515). In addition to 3 patients with intraoperative conversion to MVR, reoperation was required in 6 patients (1.4%) after successful MVP. The reason was recurrent MR due to dehiscence after resection and repair in 4 patients, infective endocarditis in 1 patient, and MS after annuloplasty with a Cosgrove 26 for

- 12 –

Hiroshi Okamoto

IMR in 1 patient. MVR was performed in 5 of these 6 patients.

Mitral valve dimensions in relation to size reproducibility, SAM, MR recurrence, and functional MS As shown in Table 3, the mean observed Lt and Lt (observed/BSA-predicted) ratio were significantly larger in the DMR group (1.8 cm, 1.06) than in the NI-FMR group (1.67 cm, 1.01) and the IMR group (1.64 cm, 0.98) (all p<0.001). Lt-predicted Dm approximated BSA-predicted Dm in the NI-FMR group (2.36, 2.35 cm) and the IMR group (2.32, 2.37 cm), while the former was larger in the DMR group (2.55, 2.37 cm). Post-repair observed Dm almost corresponded to calculated Dm (mean Dm (observed/calculated) ratio of whole group =0.99 ± 0.06). Each group was divided into three subgroups based on a Dm (observed/Lt-predicted) ratio and Dm (observed/BSA-predicted) ratio of <0.9, 0.9-1.1, or >1.1. It was found that SAM occurred in 7 patients from the DMR group with a Dm (observed/Lt-predicted) ratio <0.9 (p<0.001). In all groups, an undersized ring (Dm (observed/BSA-predicted) ratio <0.9) had no

- 13 –

Hiroshi Okamoto

beneficial effect on reducing late recurrence of MR (Table 4, Figure 3B-D). The latest mean TMPG value was the highest in the patients with a ratio <0.9. There was a significant negative correlation between the Dm (observed/BSA-predicted) ratio and the latest mean TMPG value (Figure 4) (p<0.001). The regression lines indicated that mean TMPG exceeded 5 mm Hg when the Dm (observed/BSA-predicted) ratio was <0.9 in the DMR group and when it was <0.8 in the NI-FMR and IMR groups (Figure 4).

Comment Our sizing method is technically the same as the conventional method, except we set the trigones at 1⁄4 of annular circumference instead of 1/3. Our method can be employed for complete ring sizing by using the full circle of the diagram shown in Figure 2C. In the present study, our method was used in 436 patients and we found that late control of MR among hospital survivors was comparable to that in previous reports [1, 2, 4, 5, 16, 17]. Multiple sizing techniques have been reported [6], but it seems reasonable to determine the ring size from the patient’s AML because it is the main component covering the

- 14 –

Hiroshi Okamoto

mitral orifice [1]. The conventional trigone-base sizing method assumes that the AML is a triangle with each side as Lt, and that a circle circumscribing this triangle gives the correct ring size. However, it is more anatomically rational to represent the AML as a quadrilateral because its free border is supported by the two main papillary muscles. Our research on cadaver hearts supported this concept [10]. Conventional sizing leads to 18% diameter (33% area) reduction compared with our method and may increase the risk of functional MS and SAM. AML size is primarily determined by body size (BSA) and by the degenerative process. In the absence of degenerative changes (FMR), observed Lt was almost equal to BSA-predicted Lt, while observed Lt was larger in DMR. Accordingly, Lt-predicted Dm showed the same relationship to BSA-predicted Dm (Table 3). The former parameter was related to SAM (Table 3) [18], while the latter influenced mean TMPG (Figure 4). If post-repair Dm matches Lt-predicted Dm by using our method (Figure 2), functional MS and SAM can be avoided. In addition, meticulous repair of the prolapsed leaflet in the DMR group and careful selection of IMR patients undergoing annuloplasty are essential for successful MVP.

- 15 –

Hiroshi Okamoto

Limitations The main limitations of this study were its retrospective nature and observational design. It is well known that LV remodeling can lead to a high rate of recurrent MR after MVP in IMR patients [19]. We found a slightly higher tendency for MR recurrence (p=0.052), probably because we only performed MVP for IMR patients with less tethering (tethering height ≤10 mm) [10].

Conclusions Our trigone-based sizing method, which involves setting the trigones at 1⁄4 of annular circumference, achieved satisfactory control of MR while avoiding functional MS and SAM.

- 16 –

Hiroshi Okamoto

References 1) Carpentier A. Cardiac valve surgery--the "French correction". J Thorac Cardiovasc Surg 1983; 86: 323-37. 2) David TE, Armstrong S, McCrindle BW, Manlhiot C. Late outcomes of mitral valve repair for mitral regurgitation due to degenerative disease. Circulation 2013; 127: 1485-1492 3) Mesana TG, Lam BK, Chan V, Chen K, Ruel M, Chan K. Clinical evaluation of functional mitral stenosis after mitral valve repair for degenerative disease: potential effect on surgical strategy. J Thorac Cardiovasc Surg 2013; 146: 1418-1423 4) Loulmet DFD, Yaffee W, Ursomanno PA, Rabinovich AE, Applebaum RM. Galloway AC, Grossi EA. Systolic anterior motion of the mitral valve: a 30-year perspective. J Thorac Cardiovasc Surg 2014; 148: 2787-2793. 5) Lawrie GM, Earle EA, Earle NR. Feasibility and intermediate term outcome of repair of prolapsing anterior mitral leaflets with artificial chordal replacement in 152 patients. Ann Thorac Surg 2006; 81: 849-856. 6) Bothe W, Miller DC, Doenst T. Sizing for mitral annuloplasty: where does science

- 17 –

Hiroshi Okamoto

stop and voodoo begin? Ann Thorac Surg 2013; 95: 1475-1483. 7) Gillinov AM, Cosgrove DM 3rd, Shiota T, Qin J, Tsujino H, Stewart WJ, Thomas JD, Porqueddu M, White JA, Blackstone EH. Cosgrove-Edwards Annuloplasty System: midterm results. Ann Thorac Surg 2000; 69: 717-721. 8) Okamoto H, Itoh Y, Nara Y. Geometric analysis of the anterior mitral leaflet and mitral valve orifice in cadaveric hearts. Circ J 2007; 71: 1794-99. 9) Langer F, Kunihara T, Hell K, Schramm R, Schmidt KI, Aicher D, Kindermann M, Schafers HJ. RING+STRING: Successful repair technique for ischemic mitral regurgitation with severe leaflet tethering. Circulation (2009; 120(11 Suppl): S85-91. 10) Calafiore AM, Gallina S, Di Mauro M, Gaeta F, Iacò AL, D'Alessandro S, Mazzei V, Di Giammarco G.. Mitral valve procedure in dilated cardiomyopathy: repair or replacement? Ann Thorac Surg. 2001; 71: 1146-52 11) Haycock GB, Schwartz GJ, Wisotsky DH. Geometric method for measuring body surface area: a height-weight formula validated in infants, children, and adults. J Pediatr 1978; 93: 62-66.

- 18 –

Hiroshi Okamoto

12) Prospective Studies Collaboration. Body-mass index and cause-specific mortality in 900 000 adults: collaborative analyses of 57 prospective studies. Lancet. 2009; 373:1083–1096 13) Kirklin JW, Barratt-Boyes BG. Anatomy, dimensions and terminology. In: Kirklin JW, Barratt-Boyes BG, editors. Cardiac surgery 2nd Ed, New York: Churchill Livingston; 1993. 3-60. 14) Zoghbi W. Recommendations for evaluation of the severity of native valvular regurgitation with two-dimensional and doppler echocardiography. J Amer Soc Echocard. 2003; 16: 777-802. 15) Kanda Y. Investigation of the freely available easy-to-use software 'EZR' for medical statistics. Bone Marrow Transplant 2013; 48: 452-458. 16) Brown ML, Schaff HV, Li Z, Suri RM, Daly RC, Orszulak TA. Results of mitral valve annuloplasty with a standard-sized posterior band: Is measuring important? J Thorac Cardiovasc Surg 2009; 138: 886-891 17) McNeely CA, Vassileva CM. Long-term outcomes of mitral valve repair versus replacement for degenerative disease: a systematic review. Curr Cardiol Rev 2015;

- 19 –

Hiroshi Okamoto

11(2): 157-16 18) Adams DH, Anyanwu AC, Rahmanian PB, Abascal V, Salzberg SP, Filsoufi F. Large annuloplasty rings facilitate mitral valve repair in Barlow's disease. Ann Thorac Surg 2006; 82: 2096-2100 19) Goldstein D, CTSN group. Two-year outcomes of surgical treatment of severe ischemic mitral regurgitation. N Engl J Med. 2016; 374: 344-53

- 20 –

Hiroshi Okamoto

Figure Legends Figure 1 A) Unfurled AML. Yellow quadrilateral: AML-Quad B) Sizing method. Yellow line with arrows: inter-trigonal distance (Lt) C) Implantation of an incomplete ring T: Trigones, C: commissures, Lt: inter-trigonal distance, Dm; annular diameter, APM and PPM: anterior/posterior papillary muscle attachments,

Figure 2 A) Sizer for incomplete ring B) Corresponding incomplete ring C) Diagram of the mitral annulus after incomplete ring implantation Lt: inter-trigonal distance (yellow line with arrows), Dm; annular diameter (black line with arrows), BL: band length (marker-to-marker), BW: band width, APM and PPM:

- 21 –

Hiroshi Okamoto

anterior/posterior papillary muscle attachments, Red curve with arrows: curved part of the sizer and internal border of incomplete ring (functional annulus) (these are equal in length). Dashed black curve with arrows: ring centerline (native annulus) Curved pink band: incomplete ring

Figure 3 Freedom from ≥ grade 2 recurrent MR (Kaplan-Meier analysis) A) DMR vs. NI-FMR vs. IMR, B-D); versus Dm (Ob/Pr-BSA) ratio <0.9, 0.9-1.1, or >1.1 (B: DMR, C: NI-FMR, D: IMR)

Figure 4 Scatter plot of the latest mean TMPG value versus the Dm (Ob/Pr-BSA) ratio. Black circles: DMR (D), Green crosses: NI-FMR (N), Red triangles: IMR (I) TMPG: trans-mitral pressure gradient ANCOVA: analysis of covariance - 22 –

Hiroshi Okamoto

Abbreviations AML: anterior mitral leaflet Dm: Mitral annular diameter DMR: Degenerative mitral regurgitation IMR: Ischemic mitral regurgitation Lt: Inter-trigonal distance (length) NI-FMR: Non-ischemic functional mitral regurgitation

- 23 –

Hiroshi Okamoto

Video clip; MVP for P2 prolapse

Formulae for calculation of various parameters 1) Ideal BSA (m2) =0.024265 ∗ ideal weight 0.5378 ∗ height (cm) 0.3964 [13] Ideal weight (kg) =height (m) 2 ∗ 22 [14] 2) Calculated Dm (cm) =BL ∗ 4/3 / π – BW Marker to marker band length (BL) was derived by plotting the measured BL on a Microsoft Excel chart. BW: band width, LS: labeled size. Cosgrove =0.19 ∗ LS + 0.51, (BW; 0.4) Tailor =0.18 ∗ LS + 1.34, (BW; 0.25) BSA-predicted Dm (cm) =2.0 ∗ BSA (m2) 0.36 [15] Lt-predicted Dm (cm) =observed Lt ∗ √2 3) BSA-predicted Lt =BSA-predicted Dm/√2 = √2∗ BSA (m2) 0.36

- 24 –

Hiroshi Okamoto

Table 1. Patient Characteristics & Operative results Group All DMR N 436 192 Age (y) 64.9 ± 10.2 62.6 ± 11.3 Male sex (%) 291 (67) 125 (65) BSA (m2) Ideal 1.60 ± 0.14 1.60 ± 0.15 Diabetes 130 (30) 19 (10) Atrial fibrillation 83 (19) 44 (23) NYHA class 3&4 375 (86) 141 (73) Prior cardiac surgery 13 (3) 1 (1) Labeled Band Size Cosgrove 30 (26-36) 32 (26-36) Tailor 29 (29-33) 31 (29-33) Concomitant Procedures TAP 266 (61) 155 (81) Maze 104 (24) 81 (42) CABG 190 (44) 47 (25) AVR 120 (28) 14 (7) Pump time (min) 195 ± 61 182 ± 50 ACC time (min) 105 ± 40 110 ± 35 Conversion to MVR 3 (1) 3 (2) Postoperative complications Stroke 14 (3) 3 (2) Ventilator>48 hr. 49 (11) 6 (3) Dialysis 39 (9) 3 (2) Hospital mortality 38 (8.8) 3 (1.6) NYHA class at discharge (Survivors (N=395)) * 1 380 (96) 183 (96) 2 14 (3.5) 2 (1.1) 3 1 (0.3) 1 (0.5) Late death 37 (9) 10 (5) Reoperation 9 (1.2) 8 (4.2)

NI-FMR 124 67.3 ± 9.8 70 (57) 1.57 ± 0.14 39 (31) 31 (25) 117 (94) 4 (3)

IMR 120 66.0 ± 8.0 96 (80) 1.61 ± 0.12 72 (60) 6 (5) 117 (98) 8 (7)

30 (26-34) 29 (29-31)

28 (26-34) 29 (29-31)

78 (63) 20 (16) 24 (20) 94 (76) 203 ± 68 122 ± 43 0

33(28) 3 (3) 117 (98) 12 (10) 208 ± 68 78 ± 31 0

6 (5) 20 (16) 22 (17) 18 (14.8)

5 (4) 23 (19) 14 (12) 17 (14.2)

99 (93) 7 (7) 0 10 (8) 0 (0)

98 (95) 5 (5) 0 17 (10) 1 (1)

TAP: tricuspid annuloplasty, CABG: coronary artery bypass grafting, AVR: aortic valve replacement, ACC; Aorta cross-clamp. * p< 0.001 vs. preoperative NYHA class,

- 25 –

Hiroshi Okamoto

Table 2. Echocardiography data Group All N 436

DMR 192

NI-FMR 124

IMR 120

Pre-repair LVDd (mm) LVDs (mm) LVEF (%)

SPAP (mmHg)

56 ± 8.8 39.7 ± 10.6 51.3 ± 15.9 40.5 ± 15.9

Grade of MR severity 2 11 (3) 3 225 (52) 4 200 (45) Post-repair (Pre-discharge) LVDd (mm) 48.4 ± 8.5 * LVDs (mm) 36.2 ± 9.6 * LVEF (%) 50.4 ± 13 * SPAP (mmHg) 29.3 ± 10.2 * Post-Pump Grade of MR severity * 0 393 (91) 1 35 (8) 2 5 (1) 3 3 (1) SAM 7 (1.6) Latest Grade of MR severity * 249 (63) 0 106 (27) 1 24 (6) 2 11 (3) 3 3 (1) 4 Latest TMPG (mmHg) 4.1 ± 2.1

53.7 ± 6.8 33.9 ± 6.8 62.3 ± 9.5 40.1 ± 15.9 1 (1) 34 (17) 157 (82) 45.5 ± 6.6 * 31.6 ± 6.6 * 58.8 ± 11.5 * 28.3 ± 9.6 * 175 (91) 13 (6) 1 (1) 3 (2) ** 7 (3.6) ** 110 (61) 53 (29) 13 (7) 4 (2) 1 (1) 4.2 ± 2.2

57.4 ± 10.8 41.8 ± 11.1 48.4 ± 13.5 37.8 ± 13.7

58.6 ± 8.6 47.3 ± 9.8 34.7 ± 10.8 47.2 ± 18.1

5 (4) 101 (81) 18 (15)

5 (4) 90 (75) 25 (21)

49.5 ± 9.6 * 38.1 ± 10.3 * 45.5 ± 12.3 30.8 ± 10.3 * 112 (92) 11 (8) 1 (1) 0 0 84 (78) 17 (16) 2 (2) 3 (3) 1 (1) 4.2 ± 2.1

52.5 ± 8.6 * 42.5 ± 9.4 * 45.5 ± 12.3 * 30.5 ± 11.9 * 106 (88) 11 (9) 3 (3) 0 0 55 (51) 36 (34) 11 (10) 4 (4) 1 (1) 3.8 ± 1.5

LVDd (Ds): left ventricular end-diastolic (-systolic) dimension, LVEF: left ventricular ejection fraction, SPAP: systolic pulmonary arterial pressure=Tricuspid regurgitation pressure gradient + 10, TMPG: trans-mitral pressure gradient (mean) * p<0.001: vs. Pre-repair,

** Conversion to MVR.

- 26 –

Hiroshi Okamoto

Table 3. Mitral annular diameter (Dm), inter-trigonal distance (Lt), and their ratios Group DMR NI-FMR IMR p value 114 N 182 106 Dm (Pr-BSA) (cm) 2.37 ± 0.08 2.35 ± 0.08 2.37 ± 0.06 0.070 Lt (Pr-BSA) (cm) 1.67 ± 0.06 1.66 ± 0.05 1.68 ± 0.05 0.071 (Pre-Repair) Dm (Ob) (cm) 3.72 ± 0.37 3.54 ± 0.43 Dm (Ob/Pr-BSA) ratio 1.47 ± 0.13 1.41 ± 0.14 Lt (Ob) (cm) 1.80 ± 0.20 1.67 ± 0.17 Lt (Ob/Pr-BSA) ratio 1.06 ± 0.12 1.01 ± 0.09 (D vs. N; <0.001, D vs. I; <0.001, N vs. I; 0.13) *** (Post-Repair) Dm (Ob) (cm) 2.18 ± 0.23 * 2.33 ± 0.23 * Dm (Cal) (cm) 2.36 ± 0.22 2.19 ± 0.19 Dm (Ob/Cal) ratio 0.98 ± 0.06 0.99 ± 0.06 Dm (Pr-Lt (Ob) (cm) Dm (Ob/Pr-Lt (Ob)) ratio <0.9 0.9-1.1 >1.1

Dm (Ob/Pr-BSA) ratio < 0.9 0.9-1.1 > 1.1

3.46 ± 0.29 1.36 ± 0.10 1.64 ± 0.15 0.98 ± 0.08

<0.001 <0.001 <0.001 <0.001

2.19 ± 0.19 * 2.20 ± 0.17 0.99 ± 0.05

<0.001 <0.001 0.365

2.55 ± 0.29 0.92 ± 0.08 71 (40) /7 ** 109 (60) 1 (1)

2.36 ± 0.24 0.93 ± 0.09 38 (36) 64 (60) 4 (4)

2.32 ± 0.21 0.95 ± 0.09 79 (69) 35 (61) 9 (8)

<0.001 0.014

0.98 ± 0.09 32 (18) 129 (71) 21 (12)

0.93 ± 0.09 35 (33) 70 (66) 1 (1)

0.92 ± 0.07 44 (39) 70 (61) 0

<0.001

<0.001

<0.001

Dm: Annular diameter, Ob: observed, Pr-BSA: BSA-predicted, Lt: inter-trigonal distance. Pr-Lt (Ob): Lt-predicted * p < 0.001 vs. Pre. Dm (Ob) ** Seven patients had SAM.

*** Post-hoc Bonferroni test

- 27 –

Hiroshi Okamoto

Table 4. Dm (Ob/Pr-BSA) ratio versus post-repair MR severity, SAM, and latest TMPG

Dm (Ob/Pr-BSA) ratio N

< 0.9 (A) 111

0.9 – 1.1 (B) 269

> 1.1 (C) 22

p value

Group DMR 27 (28) 129 (48) NI-FMR 35 (32) 70 (26) IMR 44 (40) 70 (26) (A vs. B; 0.005, A vs. C; <0.001, B vs. C; <0.001) *

Post-pump MR severity Grade 0 Grade 1 Grade 2 Grade 3 SAM

Latest MR severity Grade 0 Grade 1 Grade 2 Grade 3 Grade 4

21 (96) 1 (4) 0

101(91) 9 (9) 1 (1) 0 3 (2)

243 (90) 20 (7) 3 (1) 3 (1) 3 (1)

19 (86) 3 (14) 0 0 1 (4)

64 (65) 21 (21) 7 (7) 4 (4) 2 (2)

154 (62) 75 (30) 13 (5) 6 (2) 0

11 (56) 5 (28) 3 (17) 0 0

<0.001

0.836 0.328

0.130

Latest TMPG (mmHg) Mean 5.5 ± 2.8 3.6 ± 1.5 3.1 ± 1.1 <0.001 (Min-Max) (2.0-19.0) (1.0-12.0) (1.5-6.0) (A vs. B; <0.001, A vs. C; <0.001, B vs. C; 0.83) * Latest TMPG (mean) =12.3– 8.2 ∗ Dm (Ob/Pr-BSA) ratio, (R = -0.37, p<0.001) (whole group) Dm: Annular diameter, (Ob): observed, (Pr-BSA): predicted from BSA, Lt: inter-trigonal distance, SAM: systolic anterior motion, TMPG: trans-mitral pressure gradient. * Post-hoc Bonferroni test

- 28 –