Repair of Bicuspid Aortic Valves in Patients With Aortic Regurgitation
Background— Bicuspid aortic valve regurgitation can be caused by a defect in the valve itself or by dysfunction of one or more components of the aortic root complex. A successful repair thus requires correction of all aspects of the problem simultaneously. We review our experience addressing both the valve and the aortic root when correcting bicuspid valve regurgitation.
Methods and Results— Between 1996 and 2004, we treated 68 patients for aortic regurgitation. Thirty patients had isolated aortic regurgitation, and 38 had an associated ascending aortic aneurysm. All patients were treated using a standardized and integrated surgical technique, which included resection of the median raphe or leaflet plication, subcommissural annuloplasty, reinforcement of the leaflet free edge, and sinotubular junction plication. In the 38 patients with proximal aortic dilatation, reimplantation or remodeling of the aortic root was performed. Immediate postoperative echocardiography showed grade ≤1 aortic regurgitation in all patients. Three patients nonetheless needed an early re-operation because of recurrent regurgitation. No hospital mortality was observed. At a mean follow-up of 34 months after surgery, all patients were in New York Heart Association (NYHA) class 1 or 2. Two patients needed a re-operation (23 and 92 months, respectively). Echocardiographic follow-up showed no progression of the regurgitation in 58 surviving patients. Four patients progressed to grade 2 regurgitation.
Conclusion— Our data indicate that regurgitant bicuspid aortic valves, whether alone or in association with a proximal aortic dilatation, can be repaired successfully provided that both the valve and the aortic root problems are treated simultaneously.
Aortic valve replacement has been the standard surgical procedure for the treatment of aortic regurgitation since reliable prosthetic valves became available. Accordingly, in most centers, aortic valve repair has been reserved for patients with valve leakage caused by aortic disease (dissection or dilatation) or associated ventricular septal defect. In recent years, however, repair techniques for diseased aortic valves have received increasing attention. Innovations in operative techniques, an improved understanding of the functional anatomy of the aortic valve and root,1 as well as an increased awareness of the mechanisms leading to aortic regurgitation2 have undoubtedly contributed to this renewed interest.
Depending on the etiology of aortic regurgitation, several techniques can be used to repair leaking valves.3–7 Valve-preserving operations, as described by Yacoub et al4 and David et al,5,6 have been used successfully in patients with aortic regurgitation caused by diseases of the aortic root whenever the aortic leaflets were normal or nearly normal. The presence of additional cusp disease is a frequent cause of immediate failure in these operations and therefore is often considered a contraindication for this surgery. Exceptions include cases in which the underlying abnormalities are likely to be repairable,6,8 such as isolated cusp prolapse, valve fenestrations, leaflet perforation, and bicuspid aortic valves.
Several authors have examined the feasibility of repairing leaking bicuspid aortic valves, irrespective of the presence or absence of concomitant aortic root dilatation.9–12 Recently, Schaffers et al12 described an integrated functional approach that allows the surgeon to address most of the pathological aspects of the leaking bicuspid valves. Here we report on our experience with aortic valve repair.
The patient population consisted of 68 consecutive patients (65 males; mean age, 43 years; range, 16 to 76) who underwent repair of a regurgitant bicuspid aortic valve between December 1995 and November 2004. This population represents 19% of all the aortic valves repaired at our institution during the period.
All patients underwent elective surgery. Indications for surgery were the presence of severe isolated aortic regurgitation greater than grade 3 (n=27, 2 of whom had active endocarditis) and an ascending aorta aneurysm (n=36). Five patients undergoing mitral valve repair who also exhibited moderate aortic regurgitation (grade ≥2) were included as well. None of the patients had significant aortic valve stenosis. One patient had previously undergone closure of a ventricular septal defect.
Twenty-six patients were in New York Heart Association functional class 1 (38%); 28 were in class 2 (41%); 13 were in class 3 (19%); and 1 was in class 4 (2%). Three patients also had angina pectoris. The mean left ventricular ejection fraction was 59±11%. Four patients had an left ventricular ejection fraction <40%.
All patients underwent preoperative and postoperative transthoracic echocardiography as well as intraoperative transesophageal echocardiography (TEE). Transthoracic studies were performed before surgery and before hospital discharge to measure the dimensions of the ascending aorta (Table 1) and to evaluate the severity of aortic regurgitation. TEE was used to judge the adequacy of valve repair intraoperatively. Preoperatively, the severity of aortic regurgitation was graded 0 to 1 in 8 patients, 2 in 11 patients, and ≥3 in 49 patients (Table 2). The presence of a cusp prolapse was identified preoperatively in 53 patients (78%).
All patients underwent operations via median sternotomy with cardiopulmonary bypass. A transverse aortotomy was systematically performed to preserve the integrity of the aortic root and to allow for a careful inspection of all of its components. Particular attention was paid to the geometry and anatomy of both the bicuspid valve and the aortic root. The surgical strategy was tailored to the specific abnormalities seen in bicuspid valves. A standardized surgical approach addressing each of the individual components of the leaking bicuspid valve was used. Surgery included realignment of the height of the leaflets, subcommissural annuloplasty, reinforcement of the free edges of the leaflets, and plication of the sinotubular junction or root management.
Realignment of the Height of the Leaflets
For a bicuspid valve to coapt efficiently, the free edges of its leaflets must have the same length and height. The first step in valve analysis is thus to examine the length and the height of the leaflets by gently applying a radial tension on the 2 commissures and approximating the free margins. The prolapsing leaflet will be clearly identified as the one with the longest and the lower free margin. The method used thereafter to realign the leaflets depends on the geometry of the bicuspid valve.
When the leaflets are implanted symmetrically, the sinuses usually have the same depth. Correction of the prolapsing cusp can then be achieved by triangular resection or plication of the free margin (Figure 1a), depending on the quality of the leaflet tissue. This procedure is best performed using a locked running suture, supported by 2 or 3 interrupted sutures (Figure 1b). When the amount of tissue to be resected is large, as for instance in the presence of a calcified median raphe, direct closure is not advised. The gap must then be closed by either a pericardial or tricuspid valve patch.
In the presence of an asymmetrical bicuspid valve, one of the leaflets (usually the posterior one) is normal, whereas the other one is either totally or partially fused (Figure 2). The base of implantation of the posterior leaflet is also frequently smaller and lower than the anterior, and the length of the commissure is also often smaller than normal. The surgical technique used to correct the prolapsing cusp will depend on the degree of fusion between the rudimentary leaflets and the amount of excess leaflet tissue. Whenever fusion is total and leaflet tissue is in clear excess, one can use a technique similar to that described. In the presence of a partial fusion, however, a slightly different strategy should be adopted. One should first try to approximate the edges of the free margins and evaluate how this affects leaflet stiffness and mobility. One should then compare the approximated free margin to the length of the posterior leaflet to see if proper realignment can be achieved (Figure 2a). Direct approximation or closure of the partially fused leaflet with a pericardial patch should then be performed. Particular care should be taken not to reduce the mobility of the leaflets while trying to achieve the best coaptation possible.
We systematically performed subcommissural annuloplasty using an ethibond 4/0 suture secured with pledgets. This suture was routinely performed at the base of the interleaflet triangle of both commissures, as initially described by Cabrol et al.13 This suture allows reduction of the annular size, improves leaflet coaptation, and reduces the tension on the leaflet correction.
Leaflet Free Edge Reinforcement
Free margin reinforcement was performed with Gortex CV-7 (Gore W.L. Gore and Associates, Flagstaff, Ariz). For this purpose, the suture was first passed at the level of one commissure and tied. A gentle radial tension was then exerted on the 2 commissures and an over-and-over running suture was passed on each cusp and tied at the level of the other commissure, as described by David et al6 (Figure 3). Finally, a 6-0 Prolene suture was passed at the middle point of each leaflet to achieve the exact level of realignment before tying. We also used this step in patients with fragile or friable leaflet edges to strengthen the leaflet plication or repair.
Sinotubular Junction Remodeling
In patients with a normal aortic root, the sinotubular junction was systematically plicated using a pledgeted 4-0 Prolene suture placed in the middle of each sinus, at the level of the commissures. The purpose of this procedure was to maintain an ideal relationship between the annulus and the sinotubular junction.
Management of Ascending Aorta Dilatation
In patients with associated tubular ascending aneurysm, a tube graft of appropriate diameter is used to replace the aneurismal segment and thereby remodel the sinotubular junction. In case of aneurysm or dilatation of the aortic root, the techniques described were adapted for compatibility with either root remodeling or root reimplantation. The first steps always consisted of the correction of the length of the free margins, the resection of all diseased aortic tissue, and the preparation of the root for valve-sparing surgery.
In patients with asymmetrical bicuspid valves and in those with 2 rudimentary leaflets and commissures, the surgical technique was adapted to the underlying anatomy. In cases of reimplantation, the proximal suture line was always tailored to correspond to the insertion line of the leaflet, whereas in cases of remodeling, it was scalloped as described by Schaffer et al (Figure 4).12 If only one sinus was dilated, a partial remodeling was performed. Finally, when remodeling was the chosen approach, a subcommissural annuloplasty was always performed before the procedure.
After root reconstruction was completed, careful inspection of the valve was always carried out to detect any residual procedure-related cusp prolapse. Free edge reinforcement was then performed to correct for any possible differences in leaflet alignment.
After the aorta was unclamped and hemodynamic conditions had returned to normal, TEE was repeated. Any eccentric jet or aortic regurgitation (>grade 1) prompted further correction.
Thirty patients underwent an isolated bicuspid valve repair, whereas the remaining 38 underwent replacement of their ascending aorta (5 by remodeling of the sinotubular junction, 13 by remodeling of the ascending aorta, and 20 by reimplantation). Associated procedures included: mitral valve repair in 5 patients, coronary artery bypass grafting in 3 patients, aortic arch replacement in 5 patients, and closure of a patent foramen ovale in 4 patients. Surgical details are shown in Table 3.
The mean cardiopulmonary bypass time was 91 minutes (range, 66 to 194), and the mean aortic cross clamp time was 71 minutes (range, 36 to 155). Five patients (7%) needed a second pump run because of a residual aortic regurgitation (grade >1) or because of an eccentric aortic regurgitant jet. Three patients exhibited severe recurrent aortic regurgitation at days 7, 8, and 11 after surgery, respectively. All 3 patients needed re-operation. In 2, the aortic valve could be re-repaired. In the first patient, the suture supporting the median raphe resection had disrupted, and in the second, we found a suture disruption at the level of the insertion of the tricuspid autograft onto the aortic annulus. In the third, the valve could not be re-repaired, and a Ross procedure was performed. One patient needed re-operation at month 23 because of recurrent aortic regurgitation. He also underwent a Ross procedure.
In-hospital morbidity included acute myocardial infarct in 1 patient, pulmonary complications in another, and cardiac tamponade in 2 patients. No in-hospital mortality occurred. The mean postoperative hospital stay was 9 days (range, 5 to 21), which is the usual length of stay for valve surgery in our department. The discharge echocardiography showed no residual aortic regurgitation in 36 patients and aortic regurgitation grade ≤1 with a central jet in nineteen.
The mean clinical follow-up was 34 months (range, 2 to 109). Two patients died during follow-up from noncardiac causes (at 6 and 65 months, respectively) and one patient died of dissection of the ascending aorta 75 months after surgery. All survivors were in New York Heart Association classification grade 1 or 2 at the time of publication.
The mean echocardiographic follow-up was 30 months (range, 2 to 104), at which time 58 patients (85%) showed no progression of aortic regurgitation. Grade 2 aortic regurgitation was found in 4 patients. In patients with root aneurysm, the incidence of recurrent or newly developing aortic regurgitation was independent of the presence and severity of preoperative aortic regurgitation (Table 4). Finally, significant aortic stenosis developed in 1 patient, who required aortic valve replacement at month 98.
Statement of Responsibility
The authors had full access to the data and take full responsibility for their integrity. All authors have read and agree to the manuscript as written
The aim of this study was to evaluate the feasibility and the mid-term durability of bicuspid aortic valve repair using a systematic integrated approach based on a careful preoperative functional evaluation of valve pathology. Our results indicate that using such an integrated approach, successful and durable repair can be achieved in a majority of patients. Both operative and mid-term morbidity and mortality were indeed minimal, recurrence of significant aortic regurgitation was rare, and the need for reoperation was low. If we exclude the 3 patients who needed early reoperation because of immediate failure, only 2 patients needed reoperation later during follow-up. Overall, 5-year freedom from reoperation was 97%, which is somewhat better than reported in previous series. These excellent results are probably attributable to a combination of factors, which include a careful evaluation of the quality of the leaflets tissue, a detailed preoperative analysis of the mechanisms of regurgitation using TEE, the systematic use of intra-operative TEE for assessment of the results of the procedure, the use of operative technique tailored to each individual valve dysfunction, and, finally, the systematic use of root stabilization techniques to make the results more durable.
Functional Approach to Aortic Regurgitation
The management of patients with mitral regurgitation has changed considerably over the past 2 decades. Thanks to improvements in reconstructive techniques, refinements in the predictability of repair in degenerative mitral valve disease, a greater accuracy of echocardiography in defining severity and mechanisms of valve leakage, and a better understanding of the natural history of mitral insufficiency, most patients with mitral regurgitation now undergo reconstructive valve surgery rather than mitral replacement. Most are also operated on much earlier than before.14 In contrast, less progress has been made in the management of patients with aortic regurgitation. Although recent studies have demonstrated that the traditional conservative strategy of deferring aortic valve replacement until there is clear-cut evidence of ventricular dysfunction, progressive left ventricular enlargement, or symptoms resulting in excess mortality, early intervention remains the exception rather than the rule. This approach is most probably because of the fact that operation for aortic regurgitation usually leads to prosthetic replacement, which raises concerns among clinicians about the occurrence of prosthesis-related complications. These concerns would certainly be lessened if repair techniques could be used routinely in these patients as well.
The success of mitral valve repair is, without a doubt, related to the adoption of Carpentier’s functional classification of mitral valve disease.14 This classification primarily focuses on the mechanisms of valve dysfunction and provides the surgeon with a functional description of the responsible lesions that helps the surgeon in choosing the most appropriate technique to restore normal valve physiology. The same concepts can be applied to the reconstruction of the aortic valve. Recent data have indeed indicated that, like the mitral valve complex, the aortic root complex, which comprises the ventriculo-aortic junction (aortic annulus), the aortic leaflets, the sinuses of Valsalva, and the sinotubular junction, also works as an integrated functional unit to maintain aortic valve competence. In addition, increasing evidence indicates that disruption in any of the components of this functional unit can lead to the development of significant aortic regurgitation. Therefore, by analogy with Carpentier’s classification of mitral valve disease, we have recently designed a functional classification of aortic root disorders that focuses on the mechanisms of valve dysfunction as well as on a careful description of the responsible lesions. As in mitral reconstructive surgery, this classification helps surgeons choose the most appropriate techniques to repair the aortic root complex and guarantee the most durable results.
The leaking bicuspid aortic valve is a perfect illustration of the importance of this integrated functional approach. In patients with regurgitant bicuspid valves, aortic regurgitation can be caused by the simultaneous dysfunction of one or more components of the aortic root complex. A successful repair thus requires that all aspects of the problem be addressed and corrected simultaneously. As in mitral valve repair, it also requires some form of a root stabilization procedure to make the results durable.9,15 This procedure involves the ventriculo-aortic junction (aortic annulus) and the sinotubular junction, the 2 borders of the aortic root that can be considered as the equivalent of the mitral annulus.
In aortic valve repair, as in any type of valve repair, repair feasibility and long-term durability are determined primarily by the quality of the leaflet tissue and the appropriateness of the operative techniques. In each case, resection of all the diseased, calcified tissue must remain a priority and is an absolute prerequisite to any possible reconstructive strategy.
In aortic regurgitation caused by a bicuspid valve prolapse, correction of the prolapsing cusp can be achieved either by a triangular resection of the free margin or by a plication of the excess tissue.9,10 The choice between these approaches depends mainly on the quality of the leaflet tissue, with resection being the preferred option when the tissue is thickened or calcified. Plication is reserved for cases in which the leaflet tissue is still sufficiently pliable. Triangular resection is best performed leaving an edge of thickened tissue to support the suturing. The width of the resected portion should closely match that of the excess tissue to correct the prolapse. Whenever direct suturing can be performed, a locked running 6-0 Prolene suture is preferred. To avoid dehiscence, this procedure should be completed by 2 to 3 interrupted sutures, while tying the knots on the aortic surface to prevent injury to the opposite leaflet. Whenever the resected area is too large, direct suturing almost invariably results in excessive shortening of the leaflet and restriction of its movement. This result can be avoided by using a pericardial patch, the size of which should be chosen carefully to restore an appropriate leaflet length. This procedure is best performed by comparing the length of 2 leaflets while exerting a gentle radial traction on the 2 commissures. Two sutures are then passed from the edge of the resection to the corresponding point on the opposite normal leaflet (Figure 3). The same maneuver can be used in the presence of partially fused leaflets to test if direct suturing or leaflet plication is feasible or if patch extension is mandatory.
As discussed, to stabilize the repair and ensure its long-term durability, the procedure should be completed by a subcommissural annuloplasty and a plication of the sinotubular junction. By decreasing the basis of implantation of the leaflets, these maneuvers will further increase the coaptation of the leaflets as well as their motion.
Free edge reinforcement, as proposed by David et al,6 can also be used to correct a cusp prolapse. By exerting gradual tension on the suture while performing the reinforcement, one can bring the free margins to the same level above the aortic annulus. In our experience, this technique is particularly useful in the setting of valve-sparing operations. Additional benefits from free edge reinforcement are the stabilization of the free margin and the avoidance of recurrent cusp prolapse.
Patients with bicuspid aortic valves can also develop aortic regurgitation because of dilatation or dissection of the proximal aorta.2,12,16,17 In these patients, the surgical strategy should be tailored to the underlying abnormalities at either the valves or the root level. Basically, surgeons have the choice between aortic remodeling and reimplantation techniques.15–19 These techniques must of course be adapted to take into account the preexisting symmetry or asymmetry of the native bicuspid valve.12 The reimplantation technique is preferred, because it makes the procedure easier and more reproducible. The first step when using this graft is to fix the height of the native commissures at the appropriate level. This step is followed by trimming the bottom of the graft to match the patient’s anatomy. After that we have to scallop the graft in the commissure between the coronary cusp and the right to deal with the external limitations to dissection. This also helps to avoid injuring the conduction system with the sutures.
In some instances, despite these techniques, an unsuspected cusp prolapse may become apparent after reimplantation. This result is usually attributable to mild distortions of the aortic root within the graft and a resulting mismatch between the height of the prolapsing leaflet and that of the sinotubular junction. Great care should thus be taken in evaluating the relationship between these structures immediately after reimplantation. As a general rule, the level at which the aortic leaflets should come together must be at half the height of the corresponding sinuses, even when a gentle downward force is applied with a forceps as to simulate high diastolic pressures.20 Finally, any residual regurgitation (grade >1) or eccentric leak detected by TEE before coming off bypass requires prompt correction, because it will most probably lead to an early failure.
We have described an integrated and systematic approach for repairing isolated regurgitant bicuspid valves. This approach aims to correct the length of the diseased leaflet (by triangular resection or tissue plication) and to maintain the result over the long-term (by free edge reinforcement). It also aims to increase the surface of coaptation by reducing the intercommissural distance. This result is achieved usually by subcommissural annuloplasty, sinotubular plication, or both. We have also shown that this technique, together with correction of any ascending aorta dilatation, yields excellent immediate and long-term results.
Presented at the American Heart Association Scientific Sessions, Dallas, Tex, November 13–16, 2005.
Underwood MJ, El Khoury G, Glineur D, Dion R. The aortic root: structure, function, and surgical reconstruction. Heart. 2000; 83: 376–380.
Fedak PWM, Verma S, David TE, Leask RL, Weisel RD, Butany J. Clinical and pathophysiological implications of a bicuspid aortic valve. Circulation. 2002; 106: 900–904.
Haydar HS, He GW, Hovaguimian H, McUrvin D, King D, Starr A. Valve repair for aortic insufficiency: surgical classification and techniques. J Cardiothorac Surg. 1997; 11: 258–265.
Yacoub MH, Gehle P, Chandrasekaran V, Birks EJ, Child A, Radley-Smith R. Late results of a valve-preserving operation in patients with aneurysms of the ascending aorta and root. J Thoracic Cardiovasc Surg. 2001; 121: 394–403.
David TE, Ivanov J, Armstrong S, Feindel CM, Webb GD. Aortic valve-sparing operations in patients with aneurysms of the aortic root or ascending aorta. Ann Thorac Surg. 2002; 74: 1758–1761.
Westaby S, Saito S, Anastasiadis K, Moorjani N, Jin XY. Aortic root remodeling in atheromatous aneurysms: the role of selected sinus repair. Eur J Cardiothorac Surg. 2002; 21: 459–464.
Schäfers HJ, Aicher D, Langer F. Correction of leaflet prolapse in valve-preserving aortic replacement: pushing the limits? Ann Thorac Surg. 2002; 74: 1762–1764.
El Khoury G, Verhelst R, Rubay J, Noirhomme P. La réparation de la bicuspidie aortique. J Chir Thor Cardiovasc. 2000; IV: 137–140.
Casselman FP, Gillinov AM, Akhrass R, Kasirajan V, Blackstone EH, Cosgrove DM. Intermediate-term durability of bicuspid aortic valve repair for prolapsing leaflet. J Cardiothorac Surg. 1999; 15: 302–308.
Cabrol C, Guiraudon G, Bertrand M. Le traitement de l’insuffisance aortique par l’annuloplastie aortique. Arch Mal Cœur. 1966; 59: 1305–1312.
Carpentier A. Cardiac valve surgery: “the French correction”. Thorac Cardiovasc Surg. 1983; 86: 323–337.
Schaefers HJ, Fries R, Langer F, Nikoloudakis N, Graeter T, Grundmunn U. Valve-preserving replacement of the ascending aorta: remodeling versus reimplantation. J Thoracic Cardiovasc Surg. 1998; 116: 394–403.