Ataluren

Aortic root geometry following valve‑sparing root replacement with reimplantation or remodeling: experimental investigation under static continuous pressure

ImageKenichi Sasaki1,2 · Takashi Kunihara3 · Hitoshi Kasegawa4 · Masahiro Seki5 · Hiroshi Seki6 · Jumpei Takada7 · Saeko Sasuga7 · Ryo Kumazawa7 · Mitsuo Umezu2,7,8 · Kiyotaka Iwasaki2,7,8

Received: 12 October 2020 / Accepted: 23 December 2020
© The Japanese Society for Artificial Organs 2021

Abstract
The differences in aortic root geometry associated with various valve-sparing root replacement (VSRR) techniques have not fully been understood. We evaluated the root configuration of current VSRR techniques by developing in vitro test apparatus. Six fresh porcine hearts were used for each model. The aortic root remodeling control group involved replacement of the ascending aorta with diameter reduction of sino-tubular junction (STJ) (C1). The aortic valve reimplantation control group involved replacement of the ascending aorta alone (C2). VSRR included remodeling without (RM) or with annuloplasty (RM + A) and reimplantation with a tube (RI) or a handmade neo-Valsalva graft (RI + V). The root geometry of each model in response to closing hydraulic pressures of 80 and 120 mmHg was investigated using echocardiography. Among the VSRR models, RM yielded the largest aorto-ventricular junction (AVJ), which was similar to those in non-VSRR models [mean AVJ diameter (mm) at 80 mmHg; RM = 25.1 ± 1.5, RM + A = 20.9 ± 0.7, RI = 20.7 ± 0.9, RI + V = 20.8 ± 0.4]. RI + V yielded the
largest Valsalva size and largest ratio of Valsalva/AVJ, which was similar to the control group [mean Valsalva diameter (mm) at 80 mmHg; RM = 28.4 ± 1.4, RM + A = 25.8 ± 1.3, RI = 23.6 ± 1.0, RI + V = 30.5 ± 0.8, ratio of Valsalva/AVJ at 80 mmHg; RM = 1.14 ± 0.06, RM + A = 1.24 ± 0.06, RI = 1.15 ± 0.06, RI + V = 1.47 ± 0.05]. The STJ diameter at 80 mmHg was numeri-
cally smaller with RM + A (22.4 ± 1.2 mm) than with RM (24.8 ± 2.3 mm, p = 0.11). There were no significant differences in AVJ, Valsalva, or STJ distensibility or ellipticity between procedures. Current modifications, including annuloplasty for remodeling or reimplantation in the setting of neo-Valsalva graft, yield near-physiological root geometries.
Keywords Aortic root geometry · Valve-sparing root replacement · Aortic root remodeling · Aortic valve reimplantation · In vitro investigation

Supplementary Information The online version contains supplementary material available at https://doi.org/10.1007/s1004 7-020-01242-4.
* Kiyotaka Iwasaki [email protected]
1 Department of Cardiovascular Surgery, Saitama Sekishinkai Hospital, Sayama, Japan
2 Cooperative Major in Advanced Biomedical Sciences, Joint Graduate School of Tokyo Women’s Medical University and Waseda University, Waseda University, 2-2 Wakamatsucho, Shinjuku, Tokyo 162-8480, Japan
3 Department of Cardiac Surgery, The Jikei University School of Medicine, Tokyo, Japan
4 Department of Cardiac Surgery, International University of Health and Welfare, Mita Hospital, Tokyo, Japan

5 Department of Cardiac and Vascular Surgery, Dokkyo Medical University Hospital Heart Center, Tochigi, Japan
6 Division of Cardiovascular Surgery, Yamato Seiwa Hospital, Yamato, Japan
7 Department of Integrative Bioscience and Biomedical Engineering, Graduate School of Advanced Science and Engineering, Waseda University, Tokyo, Japan
8 Department of Modern Mechanical Engineering, Graduate School of Creative Science and Engineering, Waseda University, Tokyo, Japan

Introduction
Valve-sparing root replacement (VSRR) techniques have been performed to treat patients with aortic root enlarge- ment with or without aortic regurgitation (AR), especially for young active patients [1–6]. The techniques preserve native aortic valve cusps and can avoid life-long anti- coagulation therapy, which is mandatory in mechanical composite grafts. Efforts are ongoing to devise an ideal technique for avoiding reoperation and extending valve durability.

Two notable techniques, aortic root remodeling (remodeling) and aortic valve reimplantation (reim- plantation) introduced by Yacoub et al. and David et al., respectively, are widely recognized [1–3]. These origi- nal techniques had both advantages and disadvantages. However, with recent refinements, both techniques are expected to provide similar valve durability [4, 5]. The reimplantation technique fixes the aortic annulus by a vascular graft from the outside and secures annular stability. It eliminates the sinuses of Valsalva when a tube graft is employed. The valve opening and closing characteristics after reimplantation are more rapid and non-physiologic than those after remodeling [7, 8]. To overcome the non-physiologic valve dynamic behaviors observed in reimplantation with the tube graft, the neo- sinus graft has been introduced [9].
On the other hand, the remodeling technique maintains the three sinuses of Valsalva and preserves the distensi- bility of the aortic annulus during the cardiac cycles. In both in vitro and in vivo studies, remodeling exhibits more physiological valve behaviors than reimplantation. Sev- eral studies have demonstrated that annular enlargement is a risk factor for recurrent AR in remodeling procedures [10, 11]. Thus, annular-size reduction is recommended in cases with an enlarged annulus. Various approaches such as subcommissural annuloplasty, internal annuloplasty, and external annuloplasty have been advocated to prevent annular enlargement in remodeling. However, the effects of these techniques on valve behaviors are not well-under- stood in detail to date [10, 11].

The purpose of this study is to investigate the aortic root geometry in porcine aortic root models of various VSRR techniques that utilize either the straight or neo-sinus graft in reimplantation, or with or without the use of the annu- loplasty in remodeling.
Materials and methods
VSRR model

Porcine aortic roots were obtained from a local slaughter- house. The aortic roots were frozen. Then, at the day of experiments, the frozen tissues were thawed. We prepared four types of VSRR models and two control models. Six aortic valves were prepared for each model. The mod- els included remodeling with or without external suture annuloplasty and reimplantation with the tube or neo-sinus graft. We used the J Graft SHIELD NEO® (Japan Lifeline Co., Ltd, Tokyo, Japan) as the neo-sinus graft.We prepared two types of control groups for VSRR models with remodeling and reimplantation, respectively. For the two control groups, the ascending aortic replace- ment was performed with a 24 mm J-graft. As the control of remodeling (C1), the sino-tubular junction (STJ) was tied down to 22 mm by using 2–0 silk suture in order to adjust the STJ diameter between C1 and VSRR models with remodeling. As the control of reimplantation (C2), only an ascending aortic replacement with a 24 mm J-graft was performed in order to adjust the STJ diameter between C2 and VSRR models with reimplantation.

We utilized a standard technique based on a study by Schäfers et al. for the remodeling procedure [12]. We used a 22 mm tube graft of the J Graft SHIELD NEO® for all remodeling procedures. For the annuloplasty group in remodeling, a CV-0 expanded polytetrafluoroethylene suture (W.L. Gore, and Associates, Flagstaff, AZ) was used according to the report from Homburg group [13]. A 20 mm Hegar dilator was used for the annuloplasty. We define remodeling model without annuloplasty as RM, remodeling model with external suture annuloplasty as RM + A.
For the reimplantation procedure, we used a standard technique as described by David et al. in their report [3]. We made two groups using either a 24 mm tube graft of the J Graft SHIELD NEO® or a handmade neo-sinus Valsalva graft from a 24 mm J Graft SHIELD NEO®. A handmade neo-sinus Valsalva graft consists of three parts according to the native aortic root: the collar (basal ring), the sinus of Valsalva, and the straight portion (ascend- ing aorta). The direction of the groove at the collar and straight portion is horizontal, whereas at the sinus of Valsalva is vertical. The height of Valsalva was set tO 24 mm. The reasons why we used the 24 mm graft size is described as following. First, it was chosen in con- sistent with clinical usage. Second, considering an aortic wall thickness, an internal diameter after reimplantation is expected to be 4–5 mm smaller than the selected graft
size. To compare with remodeling procedure with 20 mm annuloplasty, a 24 mm graft size was chosen. We define reimplantation model of tube graft as RI, and reimplan- tation model with handmade neo-sinus Valsalva graft as RI + V.

Test apparatus and appearance of each valve-sparing root replacement model under the pressure of 80 mmHg. a Test apparatus applying static pressure on aortic valve from distal side. Valve models were placed in water chamber to measure dimensions of AVJ, Vals- alva, STJ using a ultrasonic transducer. b Appearance of each valve-
sparing root replacement model under the pressure of 80 mmHg. a Remodeling model (RM), b RM with annuloplasty (RM + A), c Reimplantation model of tube graft (RI), d Remodeling model of handmade graft with sinus (RI + V). Upper column: side views, lower column: views from the left ventricular side

Test system

We developed a test system to investigate the geometry of the aortic valves by applying constant pressure loads of each 80 and 120 mmHg from the distal side of aortic valves. The magnitudes of pressure acting on aortic valves were regulated by adjusting the height of the after-load tank. By opening the clamp between the after-load tank and valves, pressure was applied (Fig. 1a, as shown in Animation 1, 2). When the closing pressure was applied, the aortic valves were captured using a digital microscope (VHX-2000, Key- ence, Osaka, Japan). AVJ, sinus of Valsalva, and STJ were measured using the ultrasound transducer (PSM-50AT) of echocardiography (Nemio35, Toshiba, Tochigi, Japan). The measurements were conducted in the water chamber where the valve was placed. The ultrasound transducer was in direct contact with the valves. Using the areas at each portion, the diameters were calculated. The lengths of long and short axes at each portion were measured. Distensibility and ellipticity indicating characteristics of the aortic root were compared between the different VSRR techniques [7, 14–16].

The definition of distensibility indicates a relativeincrease in lumen areas due to the pressure increase. The distensibility is defined as [14, 15]:
The comparison of diameters of the AVJ, sinus of Vals- alva, and STJ between VSRR techniques and control groups are shown in Fig. 2. With regards to the AVJ, the RM group had the largest AVJ diameter among the groups except for control group. By adding annuloplasty to remodeling, the difference in AVJ diameters between the RM + A group and reimplantation groups disappeared. The sinus of the Valsalva diameter was significantly different between the groups. The Valsalva sinus of the RM and RI + V groups were larger than other groups except the control group, followed by RM + A and RI. Figure 3 shows the ratio of Valsalva/AVJ, Valsalva/ STJ, and STJ/AVJ. In the ratio of Valsalva/AVJ, RI + V had the largest value in VSRRs. By adding the suture annulo- plasty to RM, significant differences could not be achieved, but demonstrated the trend of large value in Valsalva/AVJ. With respect to the ratio of Valsalva/STJ, most of cases were more than 1.0 except for the RI group. The ratio of Valsalva/ STJ for remodeling and reimplantation groups showed sig- nificant decrease as compared with C1 and C2, respectively. Valsalva/STJ in RI were 0.97 ± 0.05 at 80 mmHg which was lower than the other VSRR groups [RM (1.16 ± 0.09), RM + A (1.16 ± 0.07), and RI + V (1.20 ± 0.11)].

As for the STJ/AVJ ratio, RI + V showed the largest value
in the VSRR groups, followed by RI, RM + A, and RM. In the VSRR groups, RM was significantly smaller than the
(maximum lumen area − minimum lumen area)∕pressure difference × minimum lumen area × 100 (%∕mmHg)
The ellipticity was examined using the following formula:
[1 − (short axis diameter∕long axis diameter)] × 100 (%)

Statistical analysis
All data were analyzed using a software package (SPSS Sta- tistics version 21, IBM, NY, USA). The distribution of nor- mality and the equality of variances were assessed according to the Shapiro–Wilk and Levene tests, respectively. Each group was analyzed using Tukey–Kramer for equal variances and Games-Howell for unequal variances. A p value of less than 0.05 was considered statistically significant.

Results
Aortic root geometry

The comparison of the macroscopic appearances of the por- cine aortic root after four types of VSRR techniques are shown in Fig. 1b.
reimplantation models, but was not smaller than C1 and C2. By adding the suture annuloplasty to RM, difference in STJ/ AVJ value between RM + A and RI disappeared.
Distensibility

The results of distensibility are shown in Fig. 4a. At the AVJ level, data showed numerical tendency that RM (20.1 ± 18.0) had the largest value, followed by C2 (19.6 ± 19.6), RI + V (18.4 ± 16.0), C1 (14.0 ± 19.1), RI (8.9 ± 7.3), and RM + A
(5.6 ± 7.0), although there were no significant differences between the groups. RM +A showed the lowest distensibility at AVJ. RI + V showed larger distensibility than RI at AVJ. At the sinus of the Valsalva level, data showed numerical tendency that RM (13.4 ± 7.7) had the largest value followed by C2 (12.0 ± 9.2), RI + V (10.6 ± 10.6), RI (8.4 ± 7.4), C1
(7.5 ± 5.9), and RM + A (7.0 ± 6.2); however, there were no significant differences between the groups. At the STJ level, RM (22.7 ± 20.1) had the largest value followed by RI + V (18.3 ± 9.8), C1 (14.8 ± 10.9), RM + A (11.5 ± 8.7),
RI (7.5 ± 6.6), and C2 (5.5 ± 3.6), but there were no sig- nificant differences between the groups. In the comparison between RM and RM + A or RI and RI + V, the trends for

Comparison of AVJ, sinus of Valsalva, and STJ diameters between valve-sparing root replacement models at 80 mmHg (the upper row) and 120 mmHg (the lower row). They are shown by box
and whisker plot, which include maximum value and minimum value, first quartile, third quartile, median and mean(blue line). **p < 0.01,
*p ≤ 0.05distensibility in Valsalva or STJ were similar to the findings at the level of AVJ.
Ellipticity

Figure 4b demonstrates the ellipticity at the AVJ level in each group. The ellipticity at the annulus for each group were as follows: RM (11.0 ± 10.4), RM + A (7.5 ± 7.7),
RI (10.4 ± 5.9), RI + V (8.9 ± 6.0), C1 (18.1 ± 7.4), C2
(18.7 ± 10.1) at 80 mmHg, RM (8.9 ± 5.2), RM + A
(9.6 ± 8.3), RI (13.5 ± 5.9), RI + V (8.7 ± 4.5), C1
(15.8 ± 12.0), and C2 (17.2 ± 9.0) at 120 mmHg. There were no significant differences between each group. However,
both remodeling and reimplantation had lower values of ellipticity than the control group.

Discussion
Our study revealed the differences in the characteristics of the root geometry between different VSRR techniques, namely remodeling with annuloplasty and reimplantation with neo-Valsalva graft. VSRR changed the diameters of AVJ, Valsalva, and STJ when compared to the control group. The Valsalva geometries were slightly different among the VSRR groups. RM, RM + A, RI, and RI + V seem bell

Comparison of ratio of Valsalva/AVJ, STJ/AVJ, and Valsalva/STJ between valve-sparing root replacement models under the pressures of 80 mmHg (the upper row) and 120 mmHg (the lower row). They are shown by box and whisker plot and mean (blue line). **p < 0.01, *p ≤ 0.05
shaped, pot shaped, straight shaped, and convex with verti- cal symmetry square shaped, respectively. As for the bulging of Valsalva sinus, namely the ratio of Valsalva/AVJ, its value was approximately 140% in RI + V, C1, C2, about 110% in RM, and nearly 120% in RM + A.
Previous observational and finite element studies have investigated an optimal geometry of the aortic root compo- nent [17–19]. Zhu et al. reported the standard human size of the proximal aorta. By using echocardiography from 314 healthy subjects, they found that the size of Valsalva sinus, STJ, and the proximal ascending aorta compared with that of aortic annulus were 142, 117, and 128% at the end-diastole and 140, 117, and 128% at the mid-systole, respectively [17]. Kazui et al. obtained similar results from normal Japanese
subjects. The size of the Valsalva sinus and STJ compared with that of the aortic annulus was 155 and 127% during the systolic phase and 155 and 123% during the diastolic phase, respectively [18]. Weltert et al. found that the optimal size of the sinuses of Valsalva through a finite element study, which contributed to the valve stress analysis and provided a desirable root size. Their research revealed that a large bulging Valsalva sinus was optimal for reducing stress. They reported that 126–140% increase in diameter of the Valsalva sinuses compared with that of STJ under an internal pressure of 200 mmHg contributed to a reduction of peak stress in the vicinity of coronary ostium [19]. Several studies sup- port the notion that approximately 140% large bulging of the Valsalva sinus for any model will be close to normal and

Comparison of distensibility and ellipticity between valve- sparing root replacement models. a Distensibility each valve-sparing root replacement model calculated using data under the pressures of
80 and 120 mmHg. b Ellipticity of each valve-sparing root replace- ment model under the pressures of 80 and 120 mmHg. They are shown by box and whisker plot and mean (blue line)has a less stressed structure. The geometries of VSSRs in our experiments were within the ranges of healthy human root geometries shown in the previous studies.

Physiological function may be related to presence of annulus stability or the Valsalva sinus rather than type of VSRR: remodeling or reimplantation. Matsumori et al. reported that the root distensibility in reimplantation improved by using a graft with sinuses through a retro- spective clinical study. They showed that the distensibility of the Valsalva sinus treated using a graft with sinuses was superior to that of a straight tube graft. The percent changes in radius (PCR) at sinus were 4.4% with a graft
with sinuses, 1.9% with a tube graft, and 7.2% in a control group. They reported a graft with sinuses compensated the sinus function in the reimplantation [16]. In our study, there were no statistical differences in root distensibility for each procedure. However, our data implied the pres- ence of relationships between the Valsalva size and disten- sibility. The RM and RI + V resulted in larger size of the Valsalva than RM + A and RI, which tend to show higher distensibility in AVJ and STJ. The higher distensibility of AVJ in RM may have an annular enlargement risk which may lead to a potential reoperation [12]. However, simi- lar to the findings of a study by Lansac et al., our results indicate that annuloplasty may decrease the risk of annular enlargement while maintaining physiological characteris- tics of remodeling [20].

Limitations
First, we used porcine aortic valves by trimming the left ven- tricle. Therefore, the aortic annulus geometries of the control groups and RM may become larger in comparison with those with the left ventricle. In addition, the porcine heart is dif- ferent from the human one in some anatomy. Ventricular septum bulging is prominent in porcine heart. It is uncertain how its thickness affects the root configuration. Second, our experiments were limited in static conditions focused on the diastolic phase. Distensibility during only the diastolic phase may differ from that during physiological pulsatile condi- tions. Investigation under pulsatile hemodynamic conditions are our future scope. Using a pulsatile circulation simula- tor, we may be able to reveal dynamic distensibility at AVJ Valsalva, and STJ between remodeling and reimplantation during systolic and diastolic phases. Third, based on a study by Graeter et al. we chose different graft sizes for remod- eling and reimplantation [8]. In most clinical cases, the graft size selection is different for the two VSRRs. Ideally, the influence of the size on root geometry should be examined. Forth, due to the limited availability, we made handmade grafts with sinuses, in which geometries are different from commercial products, and might cause larger geometrical variations. This might be related to the greater variation of distensibility in RI + V. Surgical procedural variations might exist. Indeed, although each test group mostly showed nor- mal distribution, outliers slightly existed. Fifth, it is specu- lated that local blood flow inside the Valsalva may affect the valve function. In future studies, we will plan to investigate flow in the Valsalva with various VSRR techniques using a pulsatile flow study.

Conclusion
This study revealed that remodeling without annuloplasty showed less annular stability. On the other hand, this study revealed that a suture annuloplasty strengthens the annu- lus which might decrease the risk of annular enlargement. Reimplantation with a neo-sinus graft could recreate a near physiological configuration of the sinus of Valsalva. There were no significant differences for the distensibility of the root and elliptic annular geometry among the control groups and VSRRs. Whether remodeling with annuloplasty or reim- plantation with a neo-sinus graft preserves its distensibility
under pulsatile flow conditions should be confirmed in future studies.
Acknowledgements This study was supported by Grant-in-Aid for Scientific Research (C) (No.15K10227) from Japan Society for the promotion of science (JSPS) and Subsidy Program for Development of International Standards for Evaluation of Innovative Medical Devices and Regenerative Medicine Products, from Ministry of Health, Labor and Welfare, Japan.

Author contributions KS: writing -original draft; investigation; data curation. TK: conceptualization; writing-review and editing; supervi- sion; funding acquisition. HK: conceptualization. MS: data curation; investigation. HS: data curation; investigation. SS: investigation. RK: data curation; investigation. JT: data curation; investigation. MU: con- ceptualization. KI: conceptualization; methodology; writing-review and editing; funding acquisition; supervision.

Compliance with ethical standards

Conflict of interest The author(s) declare that they have no conflicts of interest.

References
1. Yacoub MH, Fagan A, Stassano P, Radley-Smith R. Results of valve conserving operations Ataluren for aortic regurgitation. Circulation. 1983;68:311–2.
2. Sarsam MA, Yacoub M. Remodeling of the aortic valve anulus. J ThoracCardiovascSurg. 1993;105:435–8.
3. David TE, Feindel CM. An aortic valve-sparing operation for patients with aortic incompetence and aneurysm of the ascend- ing aorta. J ThoracCardiovascSurg. 1992;103:617–21.
4. Rahnavardi M, Yan TD, Bannon PG, Wilson MK. Aortic valve- sparing operations in aortic root aneurysms: remodeling or reim- plantation? Interact CardiovascThoracSurg. 2011;13:189–97.
5. Kunihara T. Valve-sparing aortic root surgery. CON: remodeling. Gen ThoracCardiovascSurg. 2019;67(1):82–92.
6. Arabkhani B, Mookhoek A, Di Centa I, Lansac E, Bekkers JA, De Lind VanWijngaarden R, et al. Reported outcome after valve-spar- ing aortic root replacement for aortic root aneurysm: a systematic review and meta-analysis. Ann Thorac Surg. 2015;100:1126–31.
7. Leyh RG, Schmidtke C, Sievers HH, Yacoub MH. Opening and closing characteristics of the aortic valve after different types of valve-preserving surgery. Circulation. 1999;100:2153–60.
8. Graeter TP, Fries R, Aicher D, Reul H, Schmitz C, Schäfers HJ. In-vitro comparison of aortic valve hemodynamics among aortic root remodeling and aortic valve reimplantation. J Heart Valve Dis. 2006;15:329–35.
9. De Paulis R, Chirichilli I, Scaffa R, Weltert L, Maselli D, Salica A, Chiariello L, et al. Long-term results of the valve reimplan- tation technique using a graft with sinuses. J ThoracCardiovasc Surg. 2016;151(1):112–9.
10. de Kerchove L, Mastrobuoni S, Boodhwani M, Astarci P, Rubay J, Poncelet A, et al. The role of annular dimension and annulo- plasty in tricuspid aortic valve repair. Eur J CardiothoracSurg. 2016;49:428–37.
11. Hanke T, Charitos EI, Stierle U, Robinson D, Gorski A, Siev- ers HH, et al. Factors associated with the development of aor- tic valve regurgitation over time after two different techniques

of valve-sparing aortic root surgery. J ThoracCardiovascSurg. 2009;137:314–9.
12. Schäfers HJ, Raddatz A, Schmied W, Takahashi H, Miura Y, Kuni- hara T, et al. Reexaminingremodeling. J ThoracCardiovascSurg. 2015;149:S30–6.
13. Schneider U, Aicher D, Miura Y, Schäfers HJ. Suture annuloplasty in aortic valve repair. Ann ThoracSurg. 2016;101:783–5.
14. Monti L, Mauri G, Balzarini L, Tarelli G, Brambilla G, Vitali E, et al. Compliance of the valsalva graft’s pseudosinuses at midterm follow-up with cardiovascular magnetic resonance. Ann Thorac- Surg. 2011;91:92–6.
15. Laurent S, Cockcroft J, Van Bortel L, Boutouyrie P, Giannattasio C, Hayoz D, et al. European network for non-invasive investiga- tion of large arteries. Expert consensus document on arterial stiff- ness: methodological issues and clinical applications. Eur Heart J. 2006;27:2588–605.
16. Matsumori M, Tanaka H, Kawanishi Y, Onishi T, Nakagiri K, Yamashita T, et al. Comparison of distensibility of the aortic root and cusp motion after aortic root replacement with two reimplan- tation techniques: valsalva graft versus tube graft. Interact Car- diovascThoracSurg. 2007;6:177–81.
17. Zhu D, Zhao Q. Dynamic normal aortic root diameters: implications for aortic root reconstruction. Ann ThoracSurg. 2011;91:485–9.
18. Kazui T, Izumoto H, Yoshioka K, Kawazoe K. Dynamic mor- phologic changes in the normal aortic annulus during systole and diastole. J H eart Valve Dis. 2006;15:617–21.
19. Weltert L, De Paulis R, Scaffa R, Maselli D, Bellisario A, D’Alessandro S. Recreation of a sinuslike graft expansion in Ben- tall procedure reduces stress at the coronary button anastomoses: a finite element study. J ThoracCardiovascSurg. 2009;137:1082–7.
20. Lansac E, Di Centa I, Sleilaty G, Lejeune S, Berrebi A, Zacek P, et al. Remodeling root repair with an external aortic ring annulo- plasty. J ThoracCardiovascSurg. 2017;153:1033–42.
Publisher’s Note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.