- Email: [email protected]
Complete replacement of the ascending aorta with reimplantation of the coronary arteries
J THoRAc
CARDIOVASC SURG
81:309-315, 1981
Current Technique
Complete replacement of the ascending aorta with reimplantation of the coronary arteries New surgical approach Thirty patients had total replacement of the ascending aorta with reimplantation of the coronary arteries, 20 for a fusiform aneurysm of the ascending aorta and 10 because of a dissection of the ascending aorta, of which three were acute. All had associated aortic insufficiency. The technique consists of implantation, within the aneurysmal sac, of a Dacron prosthesis containing a Bjork-Shiley aortic valve. The coronary orifices are anastomosed to the tubular Dacron posthesis by means of a second smaller Dacron tube. The aneurysmal pouch is then closed over the entire appliance and a fistula between the aneurysmal sac and the right atrial appendage is created to drain oozing from the prosthesis. The operative mortality was 10% (three deaths) and the late mortality has been 14.8% (four deaths). The deaths, early and late, have been confined to the first 10 cases, during which time the technique was being developed. There has been no mortality among the last 20 patients. The 23 survivors followed for an average of 19 1/2 months (range 6 months to 5 1/2 years) are in NYHA Functional Class 1 (21) or II (two). The technical modifications utilized in this series have simplified the operation and permit the proposal of this technique for aneurysms involving the entire ascending aorta.
C. Cabrol, M.D., A. Pavie, M.D., I. Gandjbakhch, M.D., J. P. Villemot, M.D.,
G. Guiraudon, M.D., L. Laughlin, M.D., Ph. Etievent, M.D., and B. Cham, M.D.,* Paris. France
Total replacement of the ascending aorta with reimplantation of the coronary arteries expands the possibilities for definitive surgical treatment of ascending aortic aneurysms with associated aortic insufficiency. Conservation or incomplete resection of the ectatic aorta leaves a persistent pathological zone which From the Service de Chirurgie Cardio- Vasculaire , Hopital La Pirie, Paris, France. Received for publication May 28, 1980. Accepted for publication July 10, 1980. Address for reprints: C. Cabrol, M.D., Service de Chirurgie Cardio- Vasculaire , Hopital La Pitie, 83, Boulevard de I 'Hopital, 75013 Paris, France. 'Current address: Department of Surgery, Lorna Linda University, Lorna Linda, Calif.
may be the site of later dissection. In the surgical technique proposed by Bentall and De Bono, 1 the coronary ostia are anastomosed directly to the Dacron prosthesis via two orifices prepared in the prosthesis. This procedure has the merit of being simple but the inconvenience of leading to traction on the fragile aortic wall adjacent to the coronary ostia, and it makes secondary hemostasis of the inferior suture line almost impossible. In the course of our experience, we have had occasion to develop some modifications of the original technique which seem to augment significantly the security of the operation. 2
Patients Thirty patients were operated upon between January, 1974, and April, 1979. There were 22 men (73.4%)
0022-5223/81/020309+07$00.70/0 © 1981 The C. V. Mosby Co.
309
3 10
Cabral et al.
The Journal of Thoracic and Cardiovascular Surgery
Fig. 1. Supravalvular aortogram. Fusiform aneurysm of the ascending aorta and aortic insufficiency.
Fig. 2. Recurrent aneurysm of the ascending aorta after partial resection of the aorta and aortic valvular replacement.
and eight women (26.6%) with an average age of 45 7/12 years (range 22 to 68). Twenty had a fusiform aneurysm of the ascending aorta (Fig. 1). Seven presented with a chronic dissection of the ascending aorta, and three had an acute dissection with origin in the ascending aorta. All had severe aortic insufficiency. Three had classic Marfan's syndrome. Five patients were in New York Heart Association (NYHA) Class II, nine were in Class III, and nine were in Class I V, including the three with acute dissections. Seven were asymptomatic. One patient presented with a recurrent aneurysm 2 years after partial resection of an ectatic ascending aorta plus aortic valve replacement (Fig. 2).
reimplanted directly on the Dacron prosthesis while the left coronary artery was reimplanted with a short Dacron tube 8 mm in diameter. The last 22 patients were operated upon according to the following technique. Following median sternotomy and evaluation of the aneurysm as well as the caliber of the aorta at the level of the brachiocephalic trunk, a Dacron prosthesis is chosen, usually 30 mm in diameter, and preclotted, either with the patients own blood prior to heparinization or with a combination of fibrinogen and thrombin. A No. 25 Bjork-Shiley aortic valve with a Pyrolite disc is sewn into the interior of the Dacron tube with a running suture of 4-0 Prolene or 3-0 Ethiftex at approximately 2 em from the proximal end after first folding back a 2 em cuff of the tube. The cuff is then replaced to its original position and the end is notched at three equidistant sites which correspond to the placement of the three aortic commissures. After systemic heparinization, extracorporeal circulation is established via the femoral artery and, aneurysmal size permitting, via the venae cavae. In the case of excessively large aneurysms which preclude catheterization of the right atrium, ve-
Surgical technique The first four patients were operated upon according to the technique described by Bentall and De Bono.! The next four were operated upon by varying techniques. In one case, the two main coronary arteries were reimplanted by means of saphenous vein grafts. In three other cases, the right coronary artery was
Volume 81 Number 2 February. 1981
Replacement of ascending aorta
3 11
Fig. 3. Operative technique. Tubular prosthesis with BjorkShiley valve plus intermediate tube anastomosed to the orifices of the coronary arteries. nous drainage is initiated via a femoral vein. Perfusion is increased progressively but with reinjection of less than the volume of venous drainage until the size of the aneurysm decreases enough to allow routine caval cannulation via the right atrium. The aorta is cross-clamped just below the brachiocephalic artery and as close to it as possible. In the case of a very fragile aorta, arterial perfusion is momentarily interrupted to permit clamping of a relaxed aortic wall. Myocardial protection is provided by a technique of profound selective myocardial hypothermia previously described." This consists of the injection of 4 L of lactated Ringer's solution at 4° C into the ascending aorta after preliminary placement of caval tapes and creation of a small opening in the right atrial wall for elimination of the Ringer's solution as it drains from the coronary sinus. Core cooling is effected to 27° C. When the intra-aortic perfusion is completed, the aneurysm is opened longitudinally and the edges are suspended by traction sutures on either side, which expose the operative site. Myocardial protection is continued by pericardial irrigation according to the technique of Shumway, Lower, and Stoffer. 4 With the edges of the aneurysm suspended, the heart can be completely submerged in the intrapericardial liquid. In general, the protection is excellent.
Fig. 4. Operative technique. Shunt from aneurysmal sac to right atrium. After examination of the lesions of the aortic wall and of the aortic valve, the latter is excised and the Dacron prosthesis is installed (Fig. 3). The end bearing the Bjork-Shiley valve is sutured to the aortic valve orifice with a running suture of 5-0 Prolene. The two coronary ostia are then anastomosed successively to the ends of a second preclotted Dacron tube 8 mm in diameter. The connection to the left coronary ostium is established initially and the Dacron tube is led circumferentially around the right flank of the aortic prosthesis anteriorly to where it is anastomosed to the right coronary ostium. The distal end of the aortic prosthesis is then sewn to the interior of the aneurysmal sac just below the origin of the brachiocephalic trunk with a continuous 4-0 Prolene suture. As with the proximal anastomosis, the running suture is started on the posterior wall and continued around both sides to be completed anteriorly. The aortic clamp is temporarily released to verify and if necessary complete hemostasis of the distal su-
The Journal of Thoracic and Cardiovascular
3 I 2 Cabrol et al.
Surgery
level of the coronary orifices, can thereby be averted. When cardiac action is satisfactory, extracorporeal perfusion is arrested and protamine is given. The oozing from suture lines stops and the connection from aorta to right atrium closes spontaneously-if not immediately, within several days.
Results
Fig. 5. Postoperative aorticangiogram showing the prosthetic aortic Dacron tube, the aortic Bjork-Shiley aortic valve, and, above, the coronary Dacron tube injecting the coronary arteries. ture line. The aorta is then reclamped and the coronary tube is connected to the aortic tube with a side-to-side anastomosis, 1.5 cm in length, on the right anterior flank of the aortic prosthesis above the Bjork-Shiley valve. The latter anastomosis is done with 5-0 Prolene suture. The two arms of the coronary prosthesis are clamped with a Kelly hemostat and the aortic crossclamp is released. Air is carefully purged from both arms of the coronary prosthesis, after which the hemostats are removed and normal coronary perfusion is resumed. The sutures are carefully examined around each coronary orifice. The patient is rewarmed and the heart defibrillates spontaneously or is easily defibrillated. When the left ventricle becomes functional and cardiac contractions become manifest on the arterial pressure curve, it is possible to test the proximal anastomosis, which is placed under tension only at that moment. It is usual, despite all precautions, to see some oozing at the level of the suture lines. The volume of oozing is variable from 100 to 600 cc/min and is sometimes not controllable before reestablishment of normal coagulation. In order to avoid prolonged aspiration of blood, the following procedure is useful. The aneurysmal wall is partially resected, but sufficient tissue is left to completely close over the entire prosthetic appliance. The periprosthetic space is then drained into the right atrium by creating a I cm fistula between that space and the right atrial appendage with a 4-0 Prolene suture" (Fig. 4). A hematoma between the prosthesis and the aneurysmal wall, which could lead to compression and eventual tearing at the
Mortality. Three of the 30 patients (10%) died within I month after operation. The 27 survivors have been followed for an average of 19V2 months (range 6 months to 5 1/ 2 years) (Fig. 5). There have been four late deaths (14.8%) during this period of follow-up. Analysis of mortality after dividing the patients into the initial 10 and the last 20 (who benefited from the above described technique) provides interesting results. All early deaths (three) occurred in the first 10 cases. Causes of death are as follows: In one of the first patients operated upon according to the technique of Bentall and De Bono, I disinsertion of the right coronary artery just at the conclusion of the procedure caused hemorrhage. A saphenous vein graft was placed from the aorta to the right coronary but it was not possible to re-establish satisfactory cardiac action. Two other patients died, one on the twelfth postoperative day as a result of septicemia (this patient had been operated upon for an acute dissection) and the second on the fifteenth postoperative day with a low cardiac output syndrome. Analysis of long-term results reveals that all late deaths (four) have occurred among the first 10 cases. Obviously, the period of observation is longer for this group, but examination of the causes of death permits several conclusions regarding the choice of therapy. The first two patients both died of an acute dissection arising at the distal anastomosis site 9 months and 41h years postoperatively. A third patient died 9 months after operation with a clinical picture of widespread myocardial infarction. Reimplantation of the coronary arteries in this patient was done by means of two saphenous vein grafts. Autopsy revealed envelopment and compression of the vein grafts in a perianeurysmal sclerotic process. This unfortunate result caused us to abandon the saphenous vein in favor of a Dacron tube for coronary artery reimplantation. One patient died at the twenty-third month of acute endocarditis, valvular thrombosis, and coronary embolus. There have been no early or late deaths among the last 20 patients. Functional results. The 23 survivors have been followed for 6 months to 5V2 years (average 19V2 months).
Volume 81 Number2 February, 1981
Replacement of ascending aorta
3 13
Fig. 6. Same case as in Fig. 5, but at a later time. Opacification of the coronary Dacron tube and the coronary arteries.
Fig. 7. Postoperative selective left coronary arteriogram showing the Dacron prosthesis connected to the orifice of the left main coronary artery.
All were re-evaluated in April, 1979, either by ourselves or by means of a questionnaire addressed to the referring physician. Twenty-one are in NYHA Class I and two are in Class II. None have presented clinical signs or electrocardiographic evidence of a coronary lesion. Four have had follow-up angiography. This small number is a reflection of the reluctance of asymptomatic patients to undergo invasive examination. Among the four, angiography reveals neither aortic nor coronary anomaly (Figs. 5 to 7).
is true, failure to excise the sinuses of Valsalva leaves the first portion of the aneurysm in place and renders the proximal anastomosis difficult by virtue of the incongruity of size between the aorta and the diameter of the prosthesis. In addition, the residual tissue is weak and the anastomosis fragile. At long term, the aneurysmal zone is subject to complications previously cited: compression, dissection, and rupture. Total replacement of the ascending aorta with reimplantation of the coronary arteries appears to be the only solution if the aneurysm involves the entire ascending aorta. This technique has permitted a significant reduction in early and late mortality, which has been observed in simple resections and partial replacements."?" Bentall' t reported satisfaction with the long-term results of his series (maximum 12 years, average 6 1/ 2 years), but the delicate portion of his technique is the reimplantation of the coronary orifices on the prosthetic aortic tube. The suturing is difficult because of imperfect visibility, fragile aortic tissue, and at times tension placed on the anastomosis. In addition, hemostatic control is quasi-impossible. These disadvantages are at the origin of immediate fatal hemorrhage or late complications such as false aneurysm at the level of coronary anastomoses, as reported by Mayer and associates l 8 and Bentall.!? Bentall's first patient died of a ruptured aneurysm at the origin of the left coronary artery. In order to avoid these complications, certain authors have envisioned the use of an intermediate tube to connect the coronary arteries and the aortic prosthesis.
Discussion Aortic insufficiency associated with ascending aortic aneurysms poses a difficult therapeutic problem. Severe aortic insufficiency indicates the need for aortic valve replacement but the route to follow vis a vis the aneurysm is more controversial. Numerous techniques have been proposed in recent years. Partial resection, either cuneiform or circumferential with end-to-end repair, is a simple technique but offers multiple disadvantages. 6, i The anastomosis is done with a certain amount of tension on a fragile aortic wall with risk of tearing after release of the aortic crossclamp. Postoperatively, the retention of pathological aortic tissue exposes the patient to the redoubtable risk of aortic rupture or acute dissection. Resection of the ascending aorta leaving the sinuses of Valsalva in place" and utilizing a Dacron prosthesis for re-establishment of continuity is a logical technique if the sinuses of Valsalva are uninvolved. If the reverse
3 14 Cabral et al.
The Journal of Thoracic and Cardiovascular Surgery
Saphenous vein grafts have been employed. 19 Our experience with the latter has been unfortunate, with progressive stenosis of the vein graft by a dense periprosthetic sclerosis. We were thus led to employ an intermediate Dacron tube, which enabled us to construct an easy anastomosis without tension and with satisfactory hemostasis. It also allowed for insertion of an aortic valvular prosthesis well above the inferior end of the aortic prosthesis, which in tum facilitated the proximal anastomosis as well as definitive hemostasis after cardiac filling. The large width of the side-to-side anastomosis between the two prosthetic tubes provides what amounts to a separate insertion of the two coronary orifices and thereby avoids the late risk of a single coronary ostium. The Dacron prosthesis is sufficiently supple to avoid kinking and adequately rigid to resist eventual compression. The closure of the ectatic aneurysmal walls around the entire prosthesis protects the appliance and allows the drainage of periprosthetic oozing into the right atrium through the fistula created between the two chambers. This eliminates trauma to red cells and plasma protein associated with prolonged pericardial aspiration and avoids the prolonged wait for complete hemostasis which at times is not crowned with success.
Conclusion These modifications of the original technique of Bentall and De Bono applied in this series have significantly decreased the mortality and morbidity in treating ascending aortic aneurysms with dilatation of the aortic anulus and aortic valvular insufficiency. They permit the selection of total replacement of the ascending aorta as the treatment of choice and avoid the secondary complications inherent in partial resectionsaortic dissection or recurrent aneurysm in the residual proximal segment. The late result of coronary reimplantation by means of an intermediary Dacron tube appears to be satisfactory for the length of follow-up observation in our series.
Addendum Since submission of the manuscript we have used this technique in operating upon 22 more patients, with one hospital death. The total number of patients so treated is 52. REFERENCES Bentall M, and De Bono A: A technique for complete replacement of ascending aorta. Thorax 23:338-339, 1968 2 Cabrol C, Gandjbakhch I, Cham B: Anevrysmes de I'aorte ascendante. Remplacement total avec reimplanta-
tion des arteres coronaires. Nouv Presse Med 7:363-365, 1978 3 Cham B, Gandjbakhch I, Guiraudon G, Cabrol C, Leclerc JC, Primo G: Hypothermie myocardique selective profonde. Nouv Presse Med 5:2327-2329,1976 4 Shumway NE, Lower R, Stoffer RC: Selective hypothermia of the heart in anoxic cardiac arrest. Surg Gynecol Obstet 109:750, 1959 5 Cabrol C: Discussion of Mayer et al" 6 Edwards WS, Kerr A: A safer technique for replacement of the entire ascending aorta and aortic valve. J THoRAc CARDIOVASC SURG 59:837-839, 1970 7 Pavie A, Gandjbakhch I, Guiraudon G, Cabrol C, et al: Anevrysmes de I'aorte ascendante avec insuffisance aortique. Coeur 9:67-77, 1978 8 Wheat MW, Wilson JR, Bartley TD: Successful replacement of the entire ascending aortia and aortic valve. JAMA 188:717-719, 1964 9 Crosby IK, Ashcraft WC, Reed WA: Surgery of proximal aorta in Marfan's syndrome. J THoRAc CARDIOVASC SURG 66:475-81,1973 10 Gandjbakhch I, Christides C, Pavie A, Mattei MF, Cabrol C, et al: Remplacement total de I'aorte ascendante avec reimplantation des arteres coronaires. Arch Mal Coeur 3:329-333, 1977 11 Helseth KM, Gaglin 11, Stentant RR, Peterson CR, Gauger DW: Ascending aortic aneurysm with associated aortic regurgitation. Ann Thorac Surg 16:368-373, 1973 12 Kouchoukos NT, Karp RB, Lell WA: Replacement of the ascending aorta and aortic valve with a composite graft. Results in 25 patients. Ann Thorac Surg 24: 140-148, 1977 13 Najafi H: Aneurysm of cystic medionecrotic aortic root. A modified surgical approach. J THORAC CARDIOVASC SURG 66:77-82, 1973 14 Rowshanzamir AM, Singh MP, Bentall M: A propos d 'une technique chirurgicale pour Ie traitement de I'anevrysme de I'aorte ascendante associe Ii une insuffisance aortique et de 11 cas operes. Chirurgie 97:704-709, 1971 15 Schulte MD, Dircks W: Aneurysm of the ascending aorta and aortic insufficiency in Marfan's syndrome. Present status of surgical management. J Cardiovasc Surg 12:469-480, 1971 16 Sing MP, Bentall M: Complete replacement of the ascending aorta and the aortic valve for the treatment of aortic aneurysm. J Cardiovasc Surg 63:218-225, 1972 17 Bentall M: Discussion of Mayer et al'" 18 Mayer JE, Lindsay WG, Wang Y, Jorgensen CR, Nicoloff DM: Composite replacement of the aortic valve and ascending aorta. J THORAC CARDIOVASC SURG 76:816-824, 1978 19 Zubiate P, Kay JH: Surgical treatment of aneurysm of the ascending aorta with aortic insufficiency and marked displacement of the coronary ostia. J THORAC CARDIOVASC SURG 71:415-421, 1976
Volume 81 Number 2 February, 1981
Reviewer's comment This is a clear and concise article with appropriate illustrations describing a new modification of the Bentall operation for aneurysmal and dissecting lesions of the ascending aorta. The authors have adopted this modified procedure and are very much encouraged with the lack of early or late mortality in their last 20 consecutive patients. I am very much impressed with their results and I agree with the authors that their modification of technique for restoration of coronary blood flow, using a separate tubular graft, does technically simplify the procedure. Despite these remarks, however, I am not certain that I would employ this technique in my next encounter with one of the lesions of the ascending aorta for which it is being proposed. On one hand our experience with the Bentall procedure with or without saphenous vein interposition has been most rewarding. Similarly, we have not had a mortality in the last 19 consecutive patients operated upon at Rush-Presbyterian-St. Luke's Medical Center. Furthermore, the two specific modifications of the Bentall operation recommended by the authors concern me. First, tubular grafts used to reconstruct the coronary arteries are potentially frought with chronic complications which can be avoided by direct implantation of the orifices of these vessels to the side of the ascending aortic graft. This concern takes on a greater significance when one considers the usually younger patients presenting with cystic medionecrotic aneurysm or dissection of the ascending aorta. I am also concerned that the wrapping of the sac of the aneurysm around the graft, a useful maneuver to contain bleeding, may cause compression, kinking, or angu-
Replacement of ascending aorta
3 15
lation of the coronary graft. This can occur both acutely and chronically as excessive cicatricial tissues may develop. Second, creation of a fistula between the perigraft space and the right atrium, although innovative, is unnecessary. Many advances in the last several years have helped in virtually eliminating the bleeding complications following the Bentall procedure. These include improved graft and suture materials; shorter periods of cardiopulmonary bypass required; markedly reduced cardiotomy suction in a relatively bloodless field and, therefore, less trauma to the blood; and finally, transfusion of fresh unrefrigerated whole blood and other coagulation blood products immediately following termination of extracorporeal circulation. My strongest objection to this report relates to the application of this radical technique in the management of acute aortic dissection. In our experience with nearly 25 such patients, a much smaller operation consisting of either primary repair or replacement of a short segment of the ascending aorta with resuspension of the prolapsed but otherwise normal aortic valve has been sufficient to rectify the dissecting process as well as to restore valve competence. We have had only one patient return 8 years after primary repair with recurrent aortic incompetence and chronic dissection who underwent a successful, classic Bentall operation. In my judgment, the operation described here should be reserved only for an exceptional situation that defies simpler surgical procedures. In closing, I wish to congratulate the authors for their superb results and their imaginative approach to the surgically demanding lesions of the ascending aorta. Hassan Najafi, M.D. Chicago, Ill.
CARDIOVASC SURG
81:309-315, 1981
Current Technique
Complete replacement of the ascending aorta with reimplantation of the coronary arteries New surgical approach Thirty patients had total replacement of the ascending aorta with reimplantation of the coronary arteries, 20 for a fusiform aneurysm of the ascending aorta and 10 because of a dissection of the ascending aorta, of which three were acute. All had associated aortic insufficiency. The technique consists of implantation, within the aneurysmal sac, of a Dacron prosthesis containing a Bjork-Shiley aortic valve. The coronary orifices are anastomosed to the tubular Dacron posthesis by means of a second smaller Dacron tube. The aneurysmal pouch is then closed over the entire appliance and a fistula between the aneurysmal sac and the right atrial appendage is created to drain oozing from the prosthesis. The operative mortality was 10% (three deaths) and the late mortality has been 14.8% (four deaths). The deaths, early and late, have been confined to the first 10 cases, during which time the technique was being developed. There has been no mortality among the last 20 patients. The 23 survivors followed for an average of 19 1/2 months (range 6 months to 5 1/2 years) are in NYHA Functional Class 1 (21) or II (two). The technical modifications utilized in this series have simplified the operation and permit the proposal of this technique for aneurysms involving the entire ascending aorta.
C. Cabrol, M.D., A. Pavie, M.D., I. Gandjbakhch, M.D., J. P. Villemot, M.D.,
G. Guiraudon, M.D., L. Laughlin, M.D., Ph. Etievent, M.D., and B. Cham, M.D.,* Paris. France
Total replacement of the ascending aorta with reimplantation of the coronary arteries expands the possibilities for definitive surgical treatment of ascending aortic aneurysms with associated aortic insufficiency. Conservation or incomplete resection of the ectatic aorta leaves a persistent pathological zone which From the Service de Chirurgie Cardio- Vasculaire , Hopital La Pirie, Paris, France. Received for publication May 28, 1980. Accepted for publication July 10, 1980. Address for reprints: C. Cabrol, M.D., Service de Chirurgie Cardio- Vasculaire , Hopital La Pitie, 83, Boulevard de I 'Hopital, 75013 Paris, France. 'Current address: Department of Surgery, Lorna Linda University, Lorna Linda, Calif.
may be the site of later dissection. In the surgical technique proposed by Bentall and De Bono, 1 the coronary ostia are anastomosed directly to the Dacron prosthesis via two orifices prepared in the prosthesis. This procedure has the merit of being simple but the inconvenience of leading to traction on the fragile aortic wall adjacent to the coronary ostia, and it makes secondary hemostasis of the inferior suture line almost impossible. In the course of our experience, we have had occasion to develop some modifications of the original technique which seem to augment significantly the security of the operation. 2
Patients Thirty patients were operated upon between January, 1974, and April, 1979. There were 22 men (73.4%)
0022-5223/81/020309+07$00.70/0 © 1981 The C. V. Mosby Co.
309
3 10
Cabral et al.
The Journal of Thoracic and Cardiovascular Surgery
Fig. 1. Supravalvular aortogram. Fusiform aneurysm of the ascending aorta and aortic insufficiency.
Fig. 2. Recurrent aneurysm of the ascending aorta after partial resection of the aorta and aortic valvular replacement.
and eight women (26.6%) with an average age of 45 7/12 years (range 22 to 68). Twenty had a fusiform aneurysm of the ascending aorta (Fig. 1). Seven presented with a chronic dissection of the ascending aorta, and three had an acute dissection with origin in the ascending aorta. All had severe aortic insufficiency. Three had classic Marfan's syndrome. Five patients were in New York Heart Association (NYHA) Class II, nine were in Class III, and nine were in Class I V, including the three with acute dissections. Seven were asymptomatic. One patient presented with a recurrent aneurysm 2 years after partial resection of an ectatic ascending aorta plus aortic valve replacement (Fig. 2).
reimplanted directly on the Dacron prosthesis while the left coronary artery was reimplanted with a short Dacron tube 8 mm in diameter. The last 22 patients were operated upon according to the following technique. Following median sternotomy and evaluation of the aneurysm as well as the caliber of the aorta at the level of the brachiocephalic trunk, a Dacron prosthesis is chosen, usually 30 mm in diameter, and preclotted, either with the patients own blood prior to heparinization or with a combination of fibrinogen and thrombin. A No. 25 Bjork-Shiley aortic valve with a Pyrolite disc is sewn into the interior of the Dacron tube with a running suture of 4-0 Prolene or 3-0 Ethiftex at approximately 2 em from the proximal end after first folding back a 2 em cuff of the tube. The cuff is then replaced to its original position and the end is notched at three equidistant sites which correspond to the placement of the three aortic commissures. After systemic heparinization, extracorporeal circulation is established via the femoral artery and, aneurysmal size permitting, via the venae cavae. In the case of excessively large aneurysms which preclude catheterization of the right atrium, ve-
Surgical technique The first four patients were operated upon according to the technique described by Bentall and De Bono.! The next four were operated upon by varying techniques. In one case, the two main coronary arteries were reimplanted by means of saphenous vein grafts. In three other cases, the right coronary artery was
Volume 81 Number 2 February. 1981
Replacement of ascending aorta
3 11
Fig. 3. Operative technique. Tubular prosthesis with BjorkShiley valve plus intermediate tube anastomosed to the orifices of the coronary arteries. nous drainage is initiated via a femoral vein. Perfusion is increased progressively but with reinjection of less than the volume of venous drainage until the size of the aneurysm decreases enough to allow routine caval cannulation via the right atrium. The aorta is cross-clamped just below the brachiocephalic artery and as close to it as possible. In the case of a very fragile aorta, arterial perfusion is momentarily interrupted to permit clamping of a relaxed aortic wall. Myocardial protection is provided by a technique of profound selective myocardial hypothermia previously described." This consists of the injection of 4 L of lactated Ringer's solution at 4° C into the ascending aorta after preliminary placement of caval tapes and creation of a small opening in the right atrial wall for elimination of the Ringer's solution as it drains from the coronary sinus. Core cooling is effected to 27° C. When the intra-aortic perfusion is completed, the aneurysm is opened longitudinally and the edges are suspended by traction sutures on either side, which expose the operative site. Myocardial protection is continued by pericardial irrigation according to the technique of Shumway, Lower, and Stoffer. 4 With the edges of the aneurysm suspended, the heart can be completely submerged in the intrapericardial liquid. In general, the protection is excellent.
Fig. 4. Operative technique. Shunt from aneurysmal sac to right atrium. After examination of the lesions of the aortic wall and of the aortic valve, the latter is excised and the Dacron prosthesis is installed (Fig. 3). The end bearing the Bjork-Shiley valve is sutured to the aortic valve orifice with a running suture of 5-0 Prolene. The two coronary ostia are then anastomosed successively to the ends of a second preclotted Dacron tube 8 mm in diameter. The connection to the left coronary ostium is established initially and the Dacron tube is led circumferentially around the right flank of the aortic prosthesis anteriorly to where it is anastomosed to the right coronary ostium. The distal end of the aortic prosthesis is then sewn to the interior of the aneurysmal sac just below the origin of the brachiocephalic trunk with a continuous 4-0 Prolene suture. As with the proximal anastomosis, the running suture is started on the posterior wall and continued around both sides to be completed anteriorly. The aortic clamp is temporarily released to verify and if necessary complete hemostasis of the distal su-
The Journal of Thoracic and Cardiovascular
3 I 2 Cabrol et al.
Surgery
level of the coronary orifices, can thereby be averted. When cardiac action is satisfactory, extracorporeal perfusion is arrested and protamine is given. The oozing from suture lines stops and the connection from aorta to right atrium closes spontaneously-if not immediately, within several days.
Results
Fig. 5. Postoperative aorticangiogram showing the prosthetic aortic Dacron tube, the aortic Bjork-Shiley aortic valve, and, above, the coronary Dacron tube injecting the coronary arteries. ture line. The aorta is then reclamped and the coronary tube is connected to the aortic tube with a side-to-side anastomosis, 1.5 cm in length, on the right anterior flank of the aortic prosthesis above the Bjork-Shiley valve. The latter anastomosis is done with 5-0 Prolene suture. The two arms of the coronary prosthesis are clamped with a Kelly hemostat and the aortic crossclamp is released. Air is carefully purged from both arms of the coronary prosthesis, after which the hemostats are removed and normal coronary perfusion is resumed. The sutures are carefully examined around each coronary orifice. The patient is rewarmed and the heart defibrillates spontaneously or is easily defibrillated. When the left ventricle becomes functional and cardiac contractions become manifest on the arterial pressure curve, it is possible to test the proximal anastomosis, which is placed under tension only at that moment. It is usual, despite all precautions, to see some oozing at the level of the suture lines. The volume of oozing is variable from 100 to 600 cc/min and is sometimes not controllable before reestablishment of normal coagulation. In order to avoid prolonged aspiration of blood, the following procedure is useful. The aneurysmal wall is partially resected, but sufficient tissue is left to completely close over the entire prosthetic appliance. The periprosthetic space is then drained into the right atrium by creating a I cm fistula between that space and the right atrial appendage with a 4-0 Prolene suture" (Fig. 4). A hematoma between the prosthesis and the aneurysmal wall, which could lead to compression and eventual tearing at the
Mortality. Three of the 30 patients (10%) died within I month after operation. The 27 survivors have been followed for an average of 19V2 months (range 6 months to 5 1/ 2 years) (Fig. 5). There have been four late deaths (14.8%) during this period of follow-up. Analysis of mortality after dividing the patients into the initial 10 and the last 20 (who benefited from the above described technique) provides interesting results. All early deaths (three) occurred in the first 10 cases. Causes of death are as follows: In one of the first patients operated upon according to the technique of Bentall and De Bono, I disinsertion of the right coronary artery just at the conclusion of the procedure caused hemorrhage. A saphenous vein graft was placed from the aorta to the right coronary but it was not possible to re-establish satisfactory cardiac action. Two other patients died, one on the twelfth postoperative day as a result of septicemia (this patient had been operated upon for an acute dissection) and the second on the fifteenth postoperative day with a low cardiac output syndrome. Analysis of long-term results reveals that all late deaths (four) have occurred among the first 10 cases. Obviously, the period of observation is longer for this group, but examination of the causes of death permits several conclusions regarding the choice of therapy. The first two patients both died of an acute dissection arising at the distal anastomosis site 9 months and 41h years postoperatively. A third patient died 9 months after operation with a clinical picture of widespread myocardial infarction. Reimplantation of the coronary arteries in this patient was done by means of two saphenous vein grafts. Autopsy revealed envelopment and compression of the vein grafts in a perianeurysmal sclerotic process. This unfortunate result caused us to abandon the saphenous vein in favor of a Dacron tube for coronary artery reimplantation. One patient died at the twenty-third month of acute endocarditis, valvular thrombosis, and coronary embolus. There have been no early or late deaths among the last 20 patients. Functional results. The 23 survivors have been followed for 6 months to 5V2 years (average 19V2 months).
Volume 81 Number2 February, 1981
Replacement of ascending aorta
3 13
Fig. 6. Same case as in Fig. 5, but at a later time. Opacification of the coronary Dacron tube and the coronary arteries.
Fig. 7. Postoperative selective left coronary arteriogram showing the Dacron prosthesis connected to the orifice of the left main coronary artery.
All were re-evaluated in April, 1979, either by ourselves or by means of a questionnaire addressed to the referring physician. Twenty-one are in NYHA Class I and two are in Class II. None have presented clinical signs or electrocardiographic evidence of a coronary lesion. Four have had follow-up angiography. This small number is a reflection of the reluctance of asymptomatic patients to undergo invasive examination. Among the four, angiography reveals neither aortic nor coronary anomaly (Figs. 5 to 7).
is true, failure to excise the sinuses of Valsalva leaves the first portion of the aneurysm in place and renders the proximal anastomosis difficult by virtue of the incongruity of size between the aorta and the diameter of the prosthesis. In addition, the residual tissue is weak and the anastomosis fragile. At long term, the aneurysmal zone is subject to complications previously cited: compression, dissection, and rupture. Total replacement of the ascending aorta with reimplantation of the coronary arteries appears to be the only solution if the aneurysm involves the entire ascending aorta. This technique has permitted a significant reduction in early and late mortality, which has been observed in simple resections and partial replacements."?" Bentall' t reported satisfaction with the long-term results of his series (maximum 12 years, average 6 1/ 2 years), but the delicate portion of his technique is the reimplantation of the coronary orifices on the prosthetic aortic tube. The suturing is difficult because of imperfect visibility, fragile aortic tissue, and at times tension placed on the anastomosis. In addition, hemostatic control is quasi-impossible. These disadvantages are at the origin of immediate fatal hemorrhage or late complications such as false aneurysm at the level of coronary anastomoses, as reported by Mayer and associates l 8 and Bentall.!? Bentall's first patient died of a ruptured aneurysm at the origin of the left coronary artery. In order to avoid these complications, certain authors have envisioned the use of an intermediate tube to connect the coronary arteries and the aortic prosthesis.
Discussion Aortic insufficiency associated with ascending aortic aneurysms poses a difficult therapeutic problem. Severe aortic insufficiency indicates the need for aortic valve replacement but the route to follow vis a vis the aneurysm is more controversial. Numerous techniques have been proposed in recent years. Partial resection, either cuneiform or circumferential with end-to-end repair, is a simple technique but offers multiple disadvantages. 6, i The anastomosis is done with a certain amount of tension on a fragile aortic wall with risk of tearing after release of the aortic crossclamp. Postoperatively, the retention of pathological aortic tissue exposes the patient to the redoubtable risk of aortic rupture or acute dissection. Resection of the ascending aorta leaving the sinuses of Valsalva in place" and utilizing a Dacron prosthesis for re-establishment of continuity is a logical technique if the sinuses of Valsalva are uninvolved. If the reverse
3 14 Cabral et al.
The Journal of Thoracic and Cardiovascular Surgery
Saphenous vein grafts have been employed. 19 Our experience with the latter has been unfortunate, with progressive stenosis of the vein graft by a dense periprosthetic sclerosis. We were thus led to employ an intermediate Dacron tube, which enabled us to construct an easy anastomosis without tension and with satisfactory hemostasis. It also allowed for insertion of an aortic valvular prosthesis well above the inferior end of the aortic prosthesis, which in tum facilitated the proximal anastomosis as well as definitive hemostasis after cardiac filling. The large width of the side-to-side anastomosis between the two prosthetic tubes provides what amounts to a separate insertion of the two coronary orifices and thereby avoids the late risk of a single coronary ostium. The Dacron prosthesis is sufficiently supple to avoid kinking and adequately rigid to resist eventual compression. The closure of the ectatic aneurysmal walls around the entire prosthesis protects the appliance and allows the drainage of periprosthetic oozing into the right atrium through the fistula created between the two chambers. This eliminates trauma to red cells and plasma protein associated with prolonged pericardial aspiration and avoids the prolonged wait for complete hemostasis which at times is not crowned with success.
Conclusion These modifications of the original technique of Bentall and De Bono applied in this series have significantly decreased the mortality and morbidity in treating ascending aortic aneurysms with dilatation of the aortic anulus and aortic valvular insufficiency. They permit the selection of total replacement of the ascending aorta as the treatment of choice and avoid the secondary complications inherent in partial resectionsaortic dissection or recurrent aneurysm in the residual proximal segment. The late result of coronary reimplantation by means of an intermediary Dacron tube appears to be satisfactory for the length of follow-up observation in our series.
Addendum Since submission of the manuscript we have used this technique in operating upon 22 more patients, with one hospital death. The total number of patients so treated is 52. REFERENCES Bentall M, and De Bono A: A technique for complete replacement of ascending aorta. Thorax 23:338-339, 1968 2 Cabrol C, Gandjbakhch I, Cham B: Anevrysmes de I'aorte ascendante. Remplacement total avec reimplanta-
tion des arteres coronaires. Nouv Presse Med 7:363-365, 1978 3 Cham B, Gandjbakhch I, Guiraudon G, Cabrol C, Leclerc JC, Primo G: Hypothermie myocardique selective profonde. Nouv Presse Med 5:2327-2329,1976 4 Shumway NE, Lower R, Stoffer RC: Selective hypothermia of the heart in anoxic cardiac arrest. Surg Gynecol Obstet 109:750, 1959 5 Cabrol C: Discussion of Mayer et al" 6 Edwards WS, Kerr A: A safer technique for replacement of the entire ascending aorta and aortic valve. J THoRAc CARDIOVASC SURG 59:837-839, 1970 7 Pavie A, Gandjbakhch I, Guiraudon G, Cabrol C, et al: Anevrysmes de I'aorte ascendante avec insuffisance aortique. Coeur 9:67-77, 1978 8 Wheat MW, Wilson JR, Bartley TD: Successful replacement of the entire ascending aortia and aortic valve. JAMA 188:717-719, 1964 9 Crosby IK, Ashcraft WC, Reed WA: Surgery of proximal aorta in Marfan's syndrome. J THoRAc CARDIOVASC SURG 66:475-81,1973 10 Gandjbakhch I, Christides C, Pavie A, Mattei MF, Cabrol C, et al: Remplacement total de I'aorte ascendante avec reimplantation des arteres coronaires. Arch Mal Coeur 3:329-333, 1977 11 Helseth KM, Gaglin 11, Stentant RR, Peterson CR, Gauger DW: Ascending aortic aneurysm with associated aortic regurgitation. Ann Thorac Surg 16:368-373, 1973 12 Kouchoukos NT, Karp RB, Lell WA: Replacement of the ascending aorta and aortic valve with a composite graft. Results in 25 patients. Ann Thorac Surg 24: 140-148, 1977 13 Najafi H: Aneurysm of cystic medionecrotic aortic root. A modified surgical approach. J THORAC CARDIOVASC SURG 66:77-82, 1973 14 Rowshanzamir AM, Singh MP, Bentall M: A propos d 'une technique chirurgicale pour Ie traitement de I'anevrysme de I'aorte ascendante associe Ii une insuffisance aortique et de 11 cas operes. Chirurgie 97:704-709, 1971 15 Schulte MD, Dircks W: Aneurysm of the ascending aorta and aortic insufficiency in Marfan's syndrome. Present status of surgical management. J Cardiovasc Surg 12:469-480, 1971 16 Sing MP, Bentall M: Complete replacement of the ascending aorta and the aortic valve for the treatment of aortic aneurysm. J Cardiovasc Surg 63:218-225, 1972 17 Bentall M: Discussion of Mayer et al'" 18 Mayer JE, Lindsay WG, Wang Y, Jorgensen CR, Nicoloff DM: Composite replacement of the aortic valve and ascending aorta. J THORAC CARDIOVASC SURG 76:816-824, 1978 19 Zubiate P, Kay JH: Surgical treatment of aneurysm of the ascending aorta with aortic insufficiency and marked displacement of the coronary ostia. J THORAC CARDIOVASC SURG 71:415-421, 1976
Volume 81 Number 2 February, 1981
Reviewer's comment This is a clear and concise article with appropriate illustrations describing a new modification of the Bentall operation for aneurysmal and dissecting lesions of the ascending aorta. The authors have adopted this modified procedure and are very much encouraged with the lack of early or late mortality in their last 20 consecutive patients. I am very much impressed with their results and I agree with the authors that their modification of technique for restoration of coronary blood flow, using a separate tubular graft, does technically simplify the procedure. Despite these remarks, however, I am not certain that I would employ this technique in my next encounter with one of the lesions of the ascending aorta for which it is being proposed. On one hand our experience with the Bentall procedure with or without saphenous vein interposition has been most rewarding. Similarly, we have not had a mortality in the last 19 consecutive patients operated upon at Rush-Presbyterian-St. Luke's Medical Center. Furthermore, the two specific modifications of the Bentall operation recommended by the authors concern me. First, tubular grafts used to reconstruct the coronary arteries are potentially frought with chronic complications which can be avoided by direct implantation of the orifices of these vessels to the side of the ascending aortic graft. This concern takes on a greater significance when one considers the usually younger patients presenting with cystic medionecrotic aneurysm or dissection of the ascending aorta. I am also concerned that the wrapping of the sac of the aneurysm around the graft, a useful maneuver to contain bleeding, may cause compression, kinking, or angu-
Replacement of ascending aorta
3 15
lation of the coronary graft. This can occur both acutely and chronically as excessive cicatricial tissues may develop. Second, creation of a fistula between the perigraft space and the right atrium, although innovative, is unnecessary. Many advances in the last several years have helped in virtually eliminating the bleeding complications following the Bentall procedure. These include improved graft and suture materials; shorter periods of cardiopulmonary bypass required; markedly reduced cardiotomy suction in a relatively bloodless field and, therefore, less trauma to the blood; and finally, transfusion of fresh unrefrigerated whole blood and other coagulation blood products immediately following termination of extracorporeal circulation. My strongest objection to this report relates to the application of this radical technique in the management of acute aortic dissection. In our experience with nearly 25 such patients, a much smaller operation consisting of either primary repair or replacement of a short segment of the ascending aorta with resuspension of the prolapsed but otherwise normal aortic valve has been sufficient to rectify the dissecting process as well as to restore valve competence. We have had only one patient return 8 years after primary repair with recurrent aortic incompetence and chronic dissection who underwent a successful, classic Bentall operation. In my judgment, the operation described here should be reserved only for an exceptional situation that defies simpler surgical procedures. In closing, I wish to congratulate the authors for their superb results and their imaginative approach to the surgically demanding lesions of the ascending aorta. Hassan Najafi, M.D. Chicago, Ill.