Citation: Boggavarapu SS, Zhao WZ, Mei KK and Feng T. Modified Bentall Procedure with Internal Tunneling Coronary Artery Transplantation.

Research Article

Modified Bentall Procedure with Internal Tunneling Coronary Artery

Transplantation

Shanmukha Sasank Boggavarapu1*#, Wen Zeng Zhao1#, Kui kui Mei1 and Tao Feng1

1Department of Cardiovascular Surgery, the First Affiliated Hospital of Zhengzhou University, PR China# Equal authorship

*Corresponding author: Shanmukha Sasank Boggavarapu, Department of Cardiovascular Surgery, The First Af-filiated Hospital of Zhengzhou University And No: 1 Jian Sha East road Zhengzhou, Henan,PR China-450000; E-mail: 991812866@qq.com

Copyright: © 2019 Shanmukha Sasank Boggavarapu

Jacobs Journal of Surgery

Abstract

Background: The Bentall procedure remains a golden standard of the aortic root surgery particularly in aortic aneurysm and ascending aortic dissection, since decades many modifications are been done to improve the outcome of the surgery. We present in our study the outcome of particular modified technique using bovine pericardium in internal tunneling coronary artery transplantation.

Methods: From 2015 March-2018 August, 26 patients (n=26) with a mean age of 52.7 ± 10.07 years went under this pro-cedure using a bovine pericardium for internal tunneling coronary artery transplantation. That helps to reduce the tension between the coronary ostium and the artificial graft and also minimize the pseudoaneurysm after surgery. All the patients were selected based on a special criterion i.e. the distance between the coronary ostium and aortic annulus is less than 2 cms. We also made some technical modifications in sewing the bovine pericardium that acts as a bib between the Dacron graft and the ostium. This technique also reduced the cross-clamp time and perfusion time.

Results: Two patients i.e. 7.6% in the group did not make it after surgery because of the myocardial failure and renal insufficiency in one patient (3.8%) and multiple organ failure in an aortic dissection patient. Low cardiac output is seen in one patient (3.8%); neurological stroke is seen in one aortic dissection patient which makes him to stay in ICU longer than other patients. Remaining 24 patients i.e. 92.4% discharged with no complications and most of them have a regular follow-up for at least 3 times in one year.

Conclusion: the modified Bentall technique with internal tunneling coronary artery transplantation has an extremely low incidence of anastomic complications particularly at the coronary ostium. Along with wrapping the ascending aorta and making a shunt to the right atrium reduced the risk of post-operative bleeding.

Keywords: Aortic aneurysm; Coronary ostium; Modified Bentall technique

Jacobs J Surg 2019; 6(4): 043.

Received Date: 19-05-2019

Accepted Date: 10-10-2019

Published Date: 06-11-2019

mailto:991812866@qq.com

2Introduction

Since the Bentall and De Bone introduced the surgi-cal procedure for the reconstruction of the aortic root with a valve composite graft [1]. This procedure has been gold-en standards for the surgical treatment for the aortic valve, root and ascending aorta [2-4]. Then later several modifica-tions have been done to improve the patient outcome and intra and post-operative complications. Here we describe a modification in Bentall procedure with internal tunneling coronary artery transplantation. Modified Bentall proce-dure with internal tunneling coronary artery transplanta-tion is done only in a special occasions where the distance between the coronary artery and aortic annulus is ≤ 2 cm. This technique have many beneficial effects like reduced tension on coronary ostium anastomosis, prevention of ex-cessive bleeding and development of false aneurysms, avoid kinking of coronary arteries , decrease cross clamp time and overall surgery time.

Figure 1: Origin of coronary arteries in different sites.

Material and Methods

This particular study was conducted in a very selec-tive group of patients who went under this bib- technique. As the patients can’t be selected preoperatively by any sort of diagnostic methods, all the patients were explained about the nature of the surgery and its possible complications and modifications that are done. A written concert was obtained from all patients.

The ethical committee of The First Affiliated Hospi-tal of Zhengzhou University also approved this study.The design of this study was approved by The First Affil-

iated Hospital of Zhengzhou University ethics committee. From March 2015 to August 2018, 26 consecutive patients underwent this modified Bentall procedure using an inter-nal tunneling coronary artery transplantation technique.

Since our institute does around 3000 surgical pro-cedures in a year, the present study was carried out in a very tiny number of patients. Still, this is a very common problem faced by most of the surgeons who are performing the Bentall procedure. The aim of the study is to prove that our technique shows good results with no pseudoaneurysm and in the ascending aorta and no kinking or tearing of the intima surrounding the coronary ostium with very good hemostatic results. Including criteria was the patients that underwent only this internal tunneling technique.

This study was approved by The First Affiliated Hos-pital of Zhengzhou University ethics committee. Patients from March 2015 to August 2018, 26 consecutive patients (21 male & 5 Female with a mean age of 57 ± 12) under-went this modified Bentall procedure using a bib technique. Patients were selected based on the size of the aortic annu-lus and the distance between the annulus and the coronary ostium ≤ 2 cm that can be observed only during surgery. The pre-operative characteristics were mentioned in table -1. Among 26 patients 23% were aortic dissection and rest 77% are shown with an aneurysm and Marfan syndrome is 3%.

Surgical technique

All the procedures were carried out through a stan-dard median sternotomy. There are various options for arterial and venous cannulation sites based on the type of dissection and the area of an aneurysm mainly on the aor-tic arch distal to the ascending aorta and proximal to the innominate artery. Alternative sites include the right sub-clavian, innominate and the femoral in case of the extended diseases. Venous cannulation is most often accomplished through the right atrium using a two-stage venous cannula-tion, while bicaval cannulation is preferred in certain cases. All surgeries were performed under moderately hypother-mic cardiopulmonary bypass (CPB). Cold blood cardiople-gia is used for myocardial protection for every 30 min and tropical cooling of heart was maintained throughout the

3procedure. The aorta was transected 1-2 cm beneath the aortic clamp. The incision was then extended towards the non-coronary sinuses in order not to damage the right cor-onary ostium. After extending the aortic root with suture the aortic valve was excised and the annulus size was mea-sured.

Internal tunneling coronary artery transplantation

According to the position and the size of the right and left coronary ostium cut a suitable size of the bovine pericardium. The left coronary ostium anastomosis if per-formed first with a bovine pericardium at the inferior bor-der (heel) with 5-0 proline sutures starting at the 3’0 clock position and continues suturing until u reach 9’0 clock posi-tion. Then leave it aside. The same is performed in the right coronary ostium.

Figure 2: Engineering the internal tunnel using bovine pericardium.

Then immediately a series of 2-0 pledged mattress suture are placed shoulder to shoulder around the aortic annulus to anchor a composite graft containing St-Jude prosthesis. Ophthalmic cautery is used to create an orifice in the graft in the appropriate position for left coronary re-implantation. The anastomosis is performed first with the bovine pericardial patch at the lower point in the inferior border of the conjugate with the bovine pericardial patch with 5-0 proline in a continuous suture manner in both di-rections till you reach 3’0 clock and 9’0 clock position. Then a knot is taken with the previous sutured proline reaches

3’0 clock and 9’0 clock position and continuous sutures were performed in an end-side anastomosis in the superior part of the coronary ostium and final knot is taken on the aortic wall. The anastomosis of the right coronary artery is performed the same way. Here we check the suture line us-ing high-pressure cardioplegia directly into the composite graft after conforming adequate hemostasis then the com-posite graft length in the measure to the distal anastomosis and performed with 4-0 continuous proline sutures. Then the composite graft is deployed and then the pericardial patch is sutured at the appropriate site with a side to side anastomosis. Before anastomosis the bovine pericardium is cut according to the length of suturing area.

Figure 3: Anastomosis of the internal tunnel to the compos-ite graft in an end- side pattern.

Wrapping

Then we use the remnant of the aortic wall to wrap around the composite graft to prevent any excess bleeding. In most of the cases, we make shunt between the aortic wall and the right atrial appendage to prevent a pseudoaneu-rysm and control bleeding. In cases like aortic dissection when the aortic wall is fragile, we use a bovine pericardium or a piece of the analogous pericardium to make a shunt. Once excellent hemostasis was obtained the patient was weaned from CPB.

4

Figure 4: Wrapping the composite graft using the remnant aortic wall.

Results and discussion

The results of most of the patients were very ex-cellent, but this is a rare type so we have only 26 samples for the following study. Here we did not compare the group with any other study. The results are described in different tables as follows pre-operative, intra operative and post-op-erative results. And follow up results were collected accord-ing to the out-patient follow up data.

Table 1: Characteristics of patients undergoing internal tunneling technique.

Sex (Male/Female) 21/5 (80.8%/19.2%)

Age (years) ± SD33-72years (52.7 ±

10.07)Pre-existing conditions n, (%)

Hypertension 19 (73%)Diabetes Mellitus 6 (23%)

Dyslipidemia 3 (11.53%)Coronary artery disease 5 (19.2%)

Chronic obstructive pulmonary disease

2 (7.6%)

Chronic kidney disease 1 (3.8%)Marfan syndrome 3 (11.53%)Smoking n, (%) 11 (42.3%)

Bicuspid aortic valve n, (%) 1 (3.8%)Previous Cardiac Surgery n, (%) 0 (%)

Aortic regurgitation 22 (84.6%)Aortic regurgitation + aortic ste-

nosis4(15.3%)

Table 2: Intraoperative data in patients.

Operation time (min) 264 ± 54

Cardiopulmonary bypass time (min) 124 ± 38

Total circulatory arrest time (min) 22 ± 18

Aortic Cross Clamp Time (min) 65 ± 27

Low cardiac output syndrome n, (%) 1 (3.8%)

Additional procedures n, (%) 3 (11.33%)

Composite valve size (mm) 23.7 ± 0.9

Post-operative bleeding

The average post-operative blood loss was 450 ± 105 mL (326 ± 95 mL in the first post-operative day and 130 ± 64 mL in the second day). Two patients (7.6%) experienced blood losses of 1000 mL or more; and in one case chest re-exploration for bleeding was required. The source of bleeding in three patients was hemorrhage from the distal tube graft anastomosis, and in the other, oozing from the proximal tube graft anastomosis was identified. In all four cases, bleeding was controlled by enforcing the anastomosis site by means of additional sutures. In the re-maining patients with blood losses of less than 1000 mL, conservative medical management in the ICU proved to be sufficient. The average number of blood unit transfusions was 2.5 (range: 1 to 4 units). However, patients who expe-rienced kidney injuries and respiratory complications re-ceived more units of blood (5.0 and 3.5 units, respectively).

Table 3: Post-operative data.

Transfusion requirements 17 (65.3%)

Reoperation for bleeding or tamponade 1(3.8%)Myocardial infarction 1(3.8%)

Neurological stroke 1(3.8%)Low cardiac output 2 (7.6%)

Renal insufficiency that require dialysis 1(3.8%)Severe respiratory complications 1(3.8%)

ECMO 1(3.8%)IABP 1(3.8%)

Complications

Acute kidney injury occurred in one patient and resolved with medical management. Neurologic sequelae were observed in one patient, one had transient ischemic

5attack. Myocardial infarction is seen in one patient that had neurological stroke. Low cardiac output in 2 patients intra-aortic balloon pump (IABP) is installed. Renal insuf-ficiency patient went on dialysis and the same patient ex-perienced severe respiratory complications that need extra corporal membrane oxygenation (ECMO) assistance. One patient after surgery died due to multi-organ failure after the surgery along with the myocardial infarction.

Short and long-term postoperative follow-up

After discharging from hospital, patients were visit-ed 30-days after discharge and annually thereafter. The fol-low-up schedule consisted of routine laboratory tests, chest radiography and echocardiography. During the follow-up period one patient died giving rise to an overall survival rate of 96.2%. The current study’s Modified Bentall proce-dure with internal tunneling coronary artery transplanta-tion believes to be simplified, thus resulting in shortening of the operation time. Simplification consists of no dissec-tion or mobilization of coronary ostia. As a dilated root is frequently associated with cephaled migration of the cor-onary ostia, approximation of the aortic wall to the graft buttons should be made directly and in front of the ostia by pre-marking the place in which the buttons should be placed. Although dissection of the ostia allows a better ex-posure and visualization of the coronary anastomosis, this could cause collateral damage, torsion of coronary ostia, pseudoaneurysms and coronary stenosis. So, the injury to the coronary ostia and the proximal coronary artery could result greater post-operative complications. However, as in this technique a great caution and patient is required to perfomre the suture line without dissection of the coronary ostia to minimize the mentioned complications. It is to note that since we do not excise the aneurysmal wall of the sinus of valsalva and sino-tubular junction, there is enough wall left that enables a proper, tensionless and well-fixed end to side suture line. So, as no tension is created, the probability of dehiscence and bleeding or pseudoaneurysm formation is lowered. As a result, operation time is shortened since minimall aortic resection and no coronary ostia dissection is required and time needed for hemostasis is minimized.

A large body of evidence over the past decades

demonstrates that for simulataneous replacement of com-posite graft of the ascending aorta and aortic valve, the orig-inal Bentall procedure and its modifications yield satisfac-tory outcome [5-9]. Studies have shown that inordinate late intraoperative and early postoperative bleeding and the for-mation of pseudoaneurysm at the suture lines are the main complications associated with the classic Bentall method [10]. It is important to reduce the operation time of the Ben-tall procedure in order to reach the minimum time for op-timum hemostasis. Previous research has explored various options and modifications in order to improve hemosta-sis and prevent bleeding. For instance, some studies have investigated the possible effects of Teflon on diminished blood loss during coronary artery anastomosis [11]. Teflon felt amplification in the aortic root has been used in acute type-A aortic dissection operation. In this approach, the Tef-lon felt was only implanted between the dissected layers of the aorta to remodel intimae layers, but was not used as the conservator layer [12]. Miller and Mitchell described the use of a doughnut of Teflon felt or autologous pericardium placed around the coronary ostial aspect of the coronary buttons to prevent tissue tearing [13]. Reinforcement with autologous pericardium during coronary artery anastomo-sis prevented late pseudoaneurysm; however, pseudoaneu-rysms of coronary ostia anastomoses were reported at the proximal aortic suture line. Recently, Della Corte et al. have suggested that in order to decrease post-operative bleed-ing, using imbricated suture-line stitches at the proximal part of vessel and spontaneous subsequent fibrin-sealant spraying were related with low rates of complications, in-cluding bleeding, renal and respiratory dysfunctions in short-term follow up with the modified Bentall procedure [14]. Besides, some studies suggested that modified Ben-tall procedure with a Carrel patch and inclusion technique could enhance hemostasis [15]. The advantages of perform-ing the Bentall procedures with a Carrel patch are that cor-onary ostial implantation into the prosthetic graft can be well executed with full visualization and that stress on the anastomosis can be averted. An open button technique is the appropriate choice for performing composite conduit replacement of the ascending aorta; however, when inser-tion of the conduit is completed an excessive post-opera-tive bleeding is still the main problem [16]. Some studies

6can solve this problem by suturing the coronary ostia in a double layer fashion with an “endo-button” technique [17], which provides a wide adherence surface against the graft and increases hemostasis. In contrast with the Teflon felt method, this technique does not need additional supportive suture by external material and as in cases of dissection; it can also be performed on the fragile aortic wall [18]. One of the remaining issues with the button technique is the injury of epicardium when the coronary buttons are going to be built. Complete coverage by the epicardium serves to con-centrate infiltration from the needle hole in the wrapping. To prevent this, the time of hemostasis and bleeding vol-ume should be controlled [19]. Cabrol technique is a safe, non-invasine and affordable technique, having a crucial role in cases of reoperation, severe calcification of the ostia, difficult mobilization of the coronary arteries and extreme aortic dilation [19]. Also, this technique could be used in coronary or aortic valve intervention in presence of chronic regeneration or familial hypercholesterolemia which leads to porcelain aorta. The technique makes the anastomoses of the coronary arteries to the aortic tube when other reim-plantation techniques fail to do so. For example, placing the brittle coronary ostia in an individual with thorough dissec-tion is difficult in the button approach, thus using the Cabrol aortocoronary anastomosis is preferred [20, 21]. It seems that the Bental operation and its Button modification have higher levels of mortality than Cabrol. Moreover, Cabrol is still considered a first-line approach in cases with specific conditions. Yet, button modification of the Bentall proce-dure remains as the standard treatment in approximately all patients [19, 22].

In the present series of patients, survival rate was 96.2%. Only 2 (7.6%) patients needed 3 days or more to stay in the ICU due to hemodynamic or respiratory prob-lems. The mean duration of hospital stay was 10 ± 2 days. On the follow-up, no cases were reported with a thrombosis or perfusion in the space between the composite graft and the coronary ostium. Post-operative complications such as myocardial infarction, endocarditis, pericardial effusion and renal failure which were related to the surgery were never reported during follow-up.

Conclusion

The modified Bentall technique with internal tun-neling coronary artery transplantation has an extremely low incidence of anastomic complications particularly at the coronary ostium; along with that there is no kinking or pseudoaneurysm formation. Wrapping the ascending aorta and making a shunt to the right atrium reduced the risk of post operative bleeding.

Ethical issue

The study was conducted in accordance with the principles of Declaration of Helsinki 1996 version and Good Practice standards. All subjects signed informed-consent forms.

Comments

Since the button technique is difficult and time con-suming and has a lower initial survival rate for the first year after operation but a better survival rate was seen in further [23].

The cabrol technique is better in redo surgeries or very complex repair such as tear in the tear of the coronary ostium which is most commonly seen in aortic dissection. Cabrol technique has a problem with the kinking of the grafts that’s connected to the coronary arteries.

The advantages of our technique are there were no tension between the coronary ostium and the composite graft and less bleeding and no pseudo aneurysm formation. We also make shunt between the aortic wall and the right atrial appendage to prevent a pseudoaneurysm and to con-trol bleeding.

References

1. Bentall H, De Bono A. A technique for complete replace-ment of the ascending aorta. Thorax 1968; 23(4):338-339.

2. Cabrol C, Pavie A, Mesnildrey P, et al. Long-term results with total replacement of the ascending aorta and reim-plantation of the coronary arteries. J Thorac Cardiovasc Surg 1986; 91(1):17-25.

3. Kouchoukos NT, Wareing TH, Murphy SF, et al. Six-teen-year experience with aortic root replacement.

https://doi.org/10.1136/thx.23.4.338https://doi.org/10.1136/thx.23.4.338https://doi.org/10.1136/thx.23.4.338https://www.ncbi.nlm.nih.gov/pubmed/2934586https://www.ncbi.nlm.nih.gov/pubmed/2934586https://www.ncbi.nlm.nih.gov/pubmed/2934586https://www.ncbi.nlm.nih.gov/pubmed/2934586https://doi.org/10.1097/00000658-199109000-00013https://doi.org/10.1097/00000658-199109000-00013

7Results of 172 operations. Ann Surg 1991; 214(3):308-318.

4. Loisance D. Simplification of the Bentall procedure with surgical and biological innovations. Eur J Cardiothorac Surg 1994; 8(11):613-615.

5. Gelsomino S, Morocutti G, Frassani R, et al. Long-term results of Bentall composite aortic root replacement for ascending aortic aneurysms and dissections. Chest 2003; 124(3):984-988.

6. Westaby S, Katsumata T, Vaccari G. Aortic root re-placement with coronary button re-implantation: low risk and predictable outcome. Eur J Cardiothorac Surg 2000; 17(3):259-265.

7. Yakut C. A new modified Bentall procedure: the flanged technique. Ann Thorac Surg 2001; 71(6):2050-2052.

8. Silva VF, Real DS, Branco JN, et al. Bentall and De Bono surgery for correction of valve and ascending aortic dis-ease: long-term results. Rev Bras Cir Cardiovasc 2008; 23(2):256-261.

9. Hashimoto W, Hashizume K, Ariyoshi T, et al. Ten years experience of aortic root replacement using a modified bentall procedure with a carrel patch and inclusion technique. Ann Vasc Dis 2011; 4(1):32-36.

10. Kouchoukos NT, Wareing TH, Murphy SF, et al. Six-teen-year experience with aortic root replacement. Results of 172 operations. Ann Surg 1991; 214(3):308-318.

11. Sokullu O, Sanioglu S, Orhan G, et al. New use of teflon to reduce bleeding in modified bentall operation. Tex Heart Inst J 2008; 35(2):147-151.

12. Bavaria JE, Pochettino A, Brinster DR, et al. New para-digms and improved results for the surgical treatment of acute type A dissection. Ann Surg 2001; 234(3):336-342.

13. Miller DC, Mitchell RS. Composite aortic valve replace-ment and graft replacement of the ascending aorta plus coronary ostial reimplantation: how I do it. Semin Tho-rac Cardiovasc Surg 1993; 5(1):74-83.

14. Della Corte A, Baldascino F, La Marca F, et al. Hemostat-ic modifications of the Bentall procedure: imbricated proximal suture and fibrin sealant reduce postoper-ative morbidity and mortality rates. Tex Heart Inst J 2012; 39(2):206-210.

15. Kawazoe K, Eishi K, Kawashima Y. New modified Bentall procedure: Carrel patch and inclusion technique. Ann Thorac Surg 1993; 55(6):1578-1579.

16. Kouchoukos NT, Marshall Jr WG, Wedige-Stecher TA. Eleven-year experience with composite graft replace-ment of the ascending aorta and aortic valve. J Thorac Cardiovasc Surg 1986; 92(4):691-705.

17. Northrup 3rd WF, Kshettry VR. Implantation technique of aortic homograft root: emphasis on matching the host root to the graft. Ann Thorac Surg 1998; 66(1):280-284.

18. Pratali S, Milano A, Codecasa R, et al. Improving hemo-stasis during replacement of the ascending aorta and aortic valve with a composite graft. Tex Heart Inst J 2000; 27(3):246-249.

19. Kourliouros A, Soni M, Rasoli S, et al. Evolution and cur-rent applications of the Cabrol procedure and its modi-fications. Ann Thorac Surg 2011; 91(5):1636-1641.

20. Grenon SM, Lachapelle K, Marcil M, et al. Surgical strat-egies for severe calcification of the aorta (porcelain aor-ta) in two patients with homozygous familial hypercho-lesterolemia. Can J Cardiol 2007; 23(14):1159-1161.

21. Mossuto E, Milano A, Bortolotti U. The Cabrol technique for replacement of the aortic valve and totally calcified ascending aorta: a case report. J Heart Valve Dis 1994; 3(1):98-100.

22. Shahian DM, O’Brien SM, Normand SL, et al. Associ-ation of hospital coronary artery bypass volume with processes of care, mortality, morbidity, and the Society of Thoracic Surgeons composite quality score. J Thorac Cardiovasc Surg 2010; 139(2):273-282.

23. Kouchoukos N.T, Wareing T.H, Murphy S.F, et al. Six-teen-year experience with aortic root replacement. Re-sults of 172 operations. Ann Surg 1991; 214: 308-320.

https://doi.org/10.1097/00000658-199109000-00013https://doi.org/10.1097/00000658-199109000-00013https://doi.org/10.1016/1010-7940(94)90046-9https://doi.org/10.1016/1010-7940(94)90046-9https://doi.org/10.1016/1010-7940(94)90046-9https://doi.org/10.1378/chest.124.3.984https://doi.org/10.1378/chest.124.3.984https://doi.org/10.1378/chest.124.3.984https://doi.org/10.1378/chest.124.3.984https://doi.org/10.1016/s1010-7940(00)00347-xhttps://doi.org/10.1016/s1010-7940(00)00347-xhttps://doi.org/10.1016/s1010-7940(00)00347-xhttps://doi.org/10.1016/s1010-7940(00)00347-xhttps://www.ncbi.nlm.nih.gov/pubmed/11426805https://www.ncbi.nlm.nih.gov/pubmed/11426805https://www.ncbi.nlm.nih.gov/pubmed/18820790https://www.ncbi.nlm.nih.gov/pubmed/18820790https://www.ncbi.nlm.nih.gov/pubmed/18820790https://www.ncbi.nlm.nih.gov/pubmed/18820790https://doi.org/10.3400/avd.oa.10.00011https://doi.org/10.3400/avd.oa.10.00011https://doi.org/10.3400/avd.oa.10.00011https://doi.org/10.3400/avd.oa.10.00011https://doi.org/10.1097/00000658-199109000-00013https://doi.org/10.1097/00000658-199109000-00013https://doi.org/10.1097/00000658-199109000-00013https://doi.org/10.1097/00000658-199109000-00013https://www.ncbi.nlm.nih.gov/pmc/articles/PMC2435449/https://www.ncbi.nlm.nih.gov/pmc/articles/PMC2435449/https://www.ncbi.nlm.nih.gov/pmc/articles/PMC2435449/https://doi.org/10.1097/00000658-200109000-00007https://doi.org/10.1097/00000658-200109000-00007https://doi.org/10.1097/00000658-200109000-00007https://doi.org/10.1097/00000658-200109000-00007https://www.ncbi.nlm.nih.gov/pubmed/8425004https://www.ncbi.nlm.nih.gov/pubmed/8425004https://www.ncbi.nlm.nih.gov/pubmed/8425004https://www.ncbi.nlm.nih.gov/pubmed/8425004https://www.ncbi.nlm.nih.gov/pubmed/22740732https://www.ncbi.nlm.nih.gov/pubmed/22740732https://www.ncbi.nlm.nih.gov/pubmed/22740732https://www.ncbi.nlm.nih.gov/pubmed/22740732https://www.ncbi.nlm.nih.gov/pubmed/22740732https://www.ncbi.nlm.nih.gov/pubmed/8512421https://www.ncbi.nlm.nih.gov/pubmed/8512421https://www.ncbi.nlm.nih.gov/pubmed/8512421https://www.ncbi.nlm.nih.gov/pubmed/3531730https://www.ncbi.nlm.nih.gov/pubmed/3531730https://www.ncbi.nlm.nih.gov/pubmed/3531730https://www.ncbi.nlm.nih.gov/pubmed/3531730https://www.ncbi.nlm.nih.gov/pubmed/9692492https://www.ncbi.nlm.nih.gov/pubmed/9692492https://www.ncbi.nlm.nih.gov/pubmed/9692492https://www.ncbi.nlm.nih.gov/pubmed/11093407https://www.ncbi.nlm.nih.gov/pubmed/11093407https://www.ncbi.nlm.nih.gov/pubmed/11093407https://www.ncbi.nlm.nih.gov/pubmed/11093407https://doi.org/10.1016/j.athoracsur.2011.01.061https://doi.org/10.1016/j.athoracsur.2011.01.061https://doi.org/10.1016/j.athoracsur.2011.01.061https://www.ncbi.nlm.nih.gov/pubmed/18060104https://www.ncbi.nlm.nih.gov/pubmed/18060104https://www.ncbi.nlm.nih.gov/pubmed/18060104https://www.ncbi.nlm.nih.gov/pubmed/18060104https://www.ncbi.nlm.nih.gov/pubmed/8162226https://www.ncbi.nlm.nih.gov/pubmed/8162226https://www.ncbi.nlm.nih.gov/pubmed/8162226https://www.ncbi.nlm.nih.gov/pubmed/8162226https://doi.org/10.1016/j.jtcvs.2009.09.007https://doi.org/10.1016/j.jtcvs.2009.09.007https://doi.org/10.1016/j.jtcvs.2009.09.007https://doi.org/10.1016/j.jtcvs.2009.09.007https://doi.org/10.1016/j.jtcvs.2009.09.007https://doi.org/10.1097/00000658-199109000-00013https://doi.org/10.1097/00000658-199109000-00013https://doi.org/10.1097/00000658-199109000-00013

_GoBack