Abstract
Objectives: In regurgitant tricuspid aortic valves, cusp prolapse may be isolated or associated with dilatation of the proximal aorta. Newly appearing cusp prolapse can also appear after an aortic valve sparing operation (AVSO) and be responsible for residual aortic regurgitation. In this report, we describe our experience in repairing prolapsing aortic cusps in 44 patients with aortic regurgitation. Methods: Between 1996 and 2003, 260 patients had aortic valve repair or valve sparing procedures in our department. All patients had peri-operative TEE. Prolapse of one or more of the aortic cusps was identified by TEE and confirmed by careful surgical inspection before and after valve sparing surgery. Forty-four patients with cusp prolapse were identified. Fifteen had an isolated prolapse, with a normal root (group I), 18 had cusp prolapse associated with dilatation of the proximal aorta (group IIa), and 11 had a newly appearing prolapse after AVSO (group IIb). Correction of the prolapsing cusp was achieved by either free edge plication, triangular resection or resuspension with PTFE. This procedure was associated with an aortic annuloplasty in group I, and with AVSO in groups II and III. Results: Post-operative TEE showed AR trivial or grade I regurgitation. At a mean of 23 months follow-up, one patient with recurrent regurgitation required an aortic valve replacement with a homograft. All remaining patients were in NYHA class I or II. Echocardiography confirmed the durability of the valve repair. Conclusions: Among the common causes of aortic regurgitation, isolated cusp prolapse is frequent and is amenable to surgical repair with excellent mid-term results. In particular, in patents who are potential candidates for AVSO, identification and correction of an associated prolapse, either pre-existing or secondary to the AVSO procedure, may further extend the indications for this technique, increase its success rates and improve its long-term outcome.
1 Introduction
Prolapse of one or more aortic cusps is a frequent cause of regurgitation. Cusp prolapse can either be isolated, with one or more of the cusps prolapsing despite a normal aortic root, be associated with dilatation of the ascending aorta, or occur as a complication of an aortic valve sparing operation (AVSO) or a stentless valve implantation. In the setting of aortic valve sparing procedures, failure to identify pre-existing or newly appearing cusp prolapses may be the cause of persisting aortic regurgitation despite an apparently successful operation.
Thanks to a better understanding of the underlying pathophysiology and recent improvements in operative techniques, aortic cusp prolapses are now considered as potentially repairable lesions [1,3–5]. In the present study, we investigated whether the systematic search, identification and subsequent surgical correction of aortic cusp prolapses, whether isolated or associated with additional aortic root pathology, allow improving the surgical outcome of aortic root surgery and increase the durability of the repair.
2 Patients and methods
Between January 1996 and July 2003, 260 patients underwent aortic valve repair or cusp preserving surgery (Table 1) . Among these patients, 44 were identified as having cusp prolapse and were included in this study. Patients were categorized into two groups. Group 1: 15 patients with isolated cusp prolapse and a normal root morphology. These patients only underwent valve repair. Group 2: 29 patients, with aneurysm of either the ascending aortic root (16 patients) or the tubular aorta (n=13), all of whom underwent aortic valve sparing surgery. Group 2 patients were further categorized into those with a pre-existing cusp prolapse (group IIa, n=18) and those in whom the cusp prolapse was only identified after the procedure (group IIb, n=11). The pre-operative characteristics of each group are shown in Table 2 . Four patients had previously undergone open heart surgery: two for a ventricular septal defect closure, one for CABG, and one for mitral valve replacement. All patients had elective surgery. No patients had aortic stenosis.
Cohort of patients who underwent aortic valve repair or valve sparing surgery in our department
Cohort of patients who underwent aortic valve repair or valve sparing surgery in our department
Preoperative patient characteristics
Preoperative patient characteristics
3 Surgical strategy
3.1 Identification of a cusp prolapse
Cusp prolapses were identified by detailed per-operative TEE examination and careful surgical inspection of the aortic valve and root. True cusp prolapse was considered to be present whenever the free margin of the cusp was below the level of its parietal insertion. A relative cusp prolapse was considered to be present whenever the free margins of each cusp were at different level, albeit above their parietal insertion.
3.1.1 TEE examination
The quality of the cusp tissue, the level of coaptation, the presence of a cusp prolapse and the overall anatomy of the aortic root were assessed in the long-axis view, while the symmetry of the aortic root and the presence of redundant tissue was evaluated in the short-axis view. Color Doppler was used to analyse the direction of the regurgitant jet within the left ventricular outflow tract (LVOT). The presence of a central jet indicates that the level of cusp coaptation is symmetrical. By contrast, eccentric jets are typically associated with asymmetrical cusp coaptation due to the presence of either a true or a relative prolapse of one or more aortic cusps.
3.1.2 Surgical inspection
A transverse aortotomy respecting the root anatomy was systematically performed. Three prolene 4/0 sutures were then passed through the three commissures. The quality of the valve tissue, the free cusp margins, as well as the anatomy of the aortic root was then evaluated. Particular attention was paid to the presence of calcifications, fenestrations or dilatation. Close inspection of the level of the free cusp margins relative to that of their parietal insertion helped to identify true prolapses. For this purpose a vertical traction was exerted on the commissures and the free-cusp edge was gently pushed in its mid-point in order to identify prolapse below the level of cusp insertion. Difference in free margin lengths, and therefore the presence of relative prolapses was estimated by applying a radial tension on the commissures. This was achieved by placing 7/0 prolene sutures through the mid-point of each cusp and applying a vertical traction. When a prolapse was present, two areas of excess cusp tissue appeared on either sides of the suture allowing to identify the prolapsing cusp.
3.2 Prolapse or length correction
The correction of relative prolapses due to unequal length of one or more cusps was achieved by using either a double radial plication of the area of excess tissue or by a single central plication, depending on the quality of the cusp tissue and the degree of prolapse. In both cases, a 6/0 running suture was used. In patients with poor tissue quality and in whom suture plication was not possible, a triangular resection was performed (Fig. 1) .
(a) Prolapse of left coronary cusp, (b) central stitch with two radial plys, showing the excess tissue, (c) plication, (d) free margin reinforcement with PTFE, (e) subcommisural annuloplasty, (f) sino-tubular junction plasty.
(a) Prolapse of left coronary cusp, (b) central stitch with two radial plys, showing the excess tissue, (c) plication, (d) free margin reinforcement with PTFE, (e) subcommisural annuloplasty, (f) sino-tubular junction plasty.
Because correction of cusp length does not automatically result in the correction of the cusp prolapse, careful inspection of the valve was always performed in order to determine whether adequate coaptation at the appropriate level had been restored. If coaptation was still inadequate or below the annular level, free edge reinforcement was applied as described below.
3.3 Subcommissural annuloplasty and free edge reinforcement
Subcommissural annuloplasty was performed using plegeted (IMPRA INC a subsidiary of CR Bard, Inc., USA) Ethibond 2/0 (Johnson and Johnson medical NV/SA) at the level of the interleaflet triangle, close to the base (Fig. 1). We used this to stabilize the aortic root and increase the area of cusp coaptation. Free margin reinforcement was performed with Gortex CV-7 (Gore W.L. Gore and Associates, Flagstaff, AZ). For this purpose, an over and over running suture was performed on each cusp while radial and vertical tensions were, respectively, exercised on the three commissures and the three margins. By exerting a gradual tension on the suture, it was possible to bring the free-cusp margins to the appropriate level, improve coaptation and correct any residual prolapse. The aim of this procedure is not only to reinforce the free margins of the cusps, but also to support the other techniques used for correction of the cusp prolapse, and hence produce a more durable repair.
3.4 Sino-tubular junction plasty
In isolated cusp prolapse associated with normal root anatomy, plasty of the sino-tubular junctions was performed using plegeted 4/0 prolene (Johnson and Johnson medical NV/SA) placed at the middle part of the inter-commissural segment (Fig. 1). Particular care was taken to avoid any overreduction which could distort the geometry of the aortic root.
3.5 Root aneurysm and pre-existing prolapse
After performing a transverse aortotomy, the cusps were carefully inspected to identify any prolapse as described earlier. Correction of the length of the cusp-free margins was always performed after preparation of the aortic valve for the sparing operation, and after excision of all diseased tissues. We used either the aortic root remodelling or the reimplantation technique. Subcommissural annuloplasty was only performed following root remodelling. After completion of the valve sparing procedure, the reconstructed aortic root was filled with the cardioplegic solution and valve competence was assessed. During this test, TEE was used to detect any residual regurgitation and, if present, to examine the direction of the jet and hence determine whether any new or residual cusp prolapse was present. After this, a gentle suction was applied superiorly to the level of cusp coaptation to further facilitate the detection of any cusp dysfunction. If prolapse was identified, it was corrected as described above. After the aorta was unclamped and haemodynamic conditions had returned back to normal, TEE was repeated. Any eccentric jet or AR grade >I prompted further correction.
4 Results
In group I, prolapse involved the right cusp in 5 patients, the noncoronary cusp in 5 other patients, the left coronary cusp in 3 additional patients and both the right and the noncoronary cusps in the last 2 patients. In group IIa (9 remodelling and 7 reimplantation), the prolapse was often difficult to detect by TEE because of the severity of aortic regurgitation. In group IIb, TEE allowed identification of the prolapse before unclamping the aorta in 5 patients and after unclamping the aorta in another 6 patients. One of the patients detected by TEE had already undergone correction of a prolapse which had been identified during the immediate post-procedural inspection. Aortic surgical procedures are shown in Table 3 .
Surgical procedures
Surgical procedures
Associated procedures were coronary bypass grafting in 9 patients, mitral valve repair in 7 patients, mitral valve replacement in 2 patients, and tricuspid valve repair in 2 patients. Mean cardiopulmonary bypass time was 78 min (range 56–165 min) in group I and 114 min (range 67–145 min) in group 2. Mean aortic cross clamp time was 57 min (range 40–123 min) in group 1 and 91 min (range 56–122) in group 2.
Four patients needed an early re-operation, 3 because of bleeding (1 in group I and 2 in group IIa) and 1 because of a recurrent aortic regurgitation at the control echocardiography on post-operative D5 (group I). At re-operation the valve could not be re-repaired and a homograft was implanted. Hospital morbidity included pneumonia in one patient, renal failure in another patient and complete A-V block in two patients. There was no hospital death. Mean post-operative hospital stay was 7±3 days (range 5–17 days), for both groups. At discharge echocardiography, 36 patients showed no residual artic regurgitation and 7 had grade I aortic regurgitation with a central jet (2 in group I, 2 in group IIa and 3 in group IIb).
Mean clinical follow-up was 23 months (range 2–84 months). One group IIa patient died at late follow-up from a cardiac cause. The other patients survived and were in NYHA class I. None needed a re-operation.
Mean echocardiographic follow-up was 15 months (range 2–50 months), and showed no progression of aortic regurgitation, except in 4 patients who developed a grade II aortic regurgitation (2 in group I, 1 in group IIa and 1 in group IIb).
5 Discussion
Aortic cusp prolapse is a frequent cause of aortic regurgitation. It can either be isolated or associated with abnormalities of the aortic root and ascending aorta. The aetiologies of isolated cusp prolapse of tricuspid aortic valves are varied (post-inflammatory degenerative, traumatic, infections, or idiopathic). Prolapse associated with aortic root pathology are usually due to the overdistension of the cusps late in the disease process or to the detachment of one or more commissures as in acute dissection. Once installed, chronic aortic regurgitation can contribute to further damage cusp structure. Turbulence from the high velocity regurgitant jet indeed result in the thickening of the free margin of the cusp, primarily in its medial aspect, while the increased stress imposed on the cusps is responsible for the fragilization of the commissural areas and the appearance of fenestrations, which in turn may ultimately rupture.
Identification of these pathological processes is important because their presence influences the surgical techniques to be used and hence the durability of the repair. We have indeed learnt from mitral repair [12] that there are two important determinants of repair feasibility and durability: the quality of valve tissue left behind after surgery and the use of appropriate surgical technique[s].
Initial techniques for correction of aortic cusp prolapse have been described by Carpentier [12] and Duran et al. [2,4]. These authors reported excellent early results in patients whom prolapse aetiology was mainly rheumatic [4] or congenital [13] in origin. Long-term results have been also reported by Casselman et al. [5] in bicuspid aortic valves as well as after traumatic disruption of tricuspid aortic valves [14]. In these studies, repair failure, whenever occurring, was attributed to either progression of the underlying disease, recurrence of the cusp prolapse, dilation of the aorto-ventricular and sino-tubular junctions or inadequate surgical techniques.
Currently, there are no data about the long-term results of repair procedures to treat patients with a degenerative prolapse of a tricuspid aortic valve. Since tissue quality is usually good in these patients, we sought to investigate whether degenerative cusp prolapses of tricuspid aortic valves can be repaired by using surgical techniques that restore cusp coaptation and stabilize the aortic root.
5.1 Surgical considerations
Careful surgical inspection, together with a detailed per-operative TEE analysis, enabled us to identify cusp prolapses in a significant number of patients with isolated aortic regurgitation. It also allowed us to determine which cusp was involved and whether we were dealing with a true or a relative cusp prolapse. To make this difference, one needs to define an adequate reference point to which the actual level of cusp coaptation can be compared. When only one cusp is prolapsing, this can readily be achieved by taking the other two cusps as reference levels. When all three cusps are prolapsing, which is rare, the appropriate coaptation level is more difficult to define. In this case, we usually considered that the optimum height for coaptation should be the middle distance of the height of the commissure.
From a technical perspective, correction of the prolapse was most often achieved by using either a double radial plication of the area of excess tissue or by a single central plication, depending on the quality of the cusp tissue and the degree of prolapse. In patients with poor tissue quality and in whom suture plication was not possible, a triangular resection was performed. We never used commissural re-suspension or plication near the commissure because of the fragile nature of the tissue at this level and the presence of associated fenestrations. Unfortunately, correction of the free margin length by itself was frequently not sufficient to cure the prolapse. Indeed, at the end of this procedure, many patients still demonstrate a rather small coaptation area. Subcommisural anuloplasty helped us to increase the coaptation area. We always combined it with a free edge reinforcement to heighten the shortened cusp if necessary, to further stabilize the more fragile segments of the valve and reduce the incidence of prolapse recurrence.
Restoration of a normal sino-tubular junction is another important aspect of our repair strategy. In patients with tubular aneurysm, this is best achieved by replacing the tubular aorta, and adjusting the diameter of the graft to the diameter of the neo aorto-ventricular diameter (post-subcommisural annuloplasty). In patients with a normal tubular aorta, similar results can be obtained by performing three plications in the middle point of each sino-tubular junction segment. This further helps stabilizing the cusp repair and the aortic root itself, in much the same way correction of mitral leaflet prolapse and annular dilatation help stabilizing mitral reconstructions and increase the durability of the repair.
Aortic valve sparing surgery has been shown to be an excellent option for patients with proximal aortic aneurysm with or without aortic regurgitation. Two operative techniques have been described: aortic root remodelling [9,15] and reimplantation [8,15]. Although good results have been reported by several groups [6–11], many surgeons have abandoned these procedures because of a high incidence of immediate or early repair failures and the frequent persistence of severe aortic regurgitation after the operation. Although the exact reason for these failures remains a matter of debate, we believe that they are most likely due to either the presence of a pre-existing but unrecognized cusp prolapse or the appearance of a new prolapse, as a result of the surgical procedure itself. Asymmetrical re-approximation and overreduction of the sino-tubular junction have indeed been shown to cause cusp prolapse [16]. This was the case in two patients after a Ross procedure in which overreduction of the sino-tubular junction resulted in cusp prolapse and severe aortic regurgitation. As in cases of isolated cusp prolapse, careful analysis of the cups by TEE and by surgical inspection may help to identify these cusp prolapses and should prompt the surgeons to apply the appropriate surgical strategy. In this respect, our data indicate that whenever a cusp prolapse has been identified during a valve sparing operation, its correction allows to decrease the number of early failures in patients and improve long-term durability.
6 Conclusion
Our strategy of repairing aortic valve prolapse and stabilizing the supporting structures of the aortic root seems promising. Systematic detection of cusp prolapse in complex cases of aortic root reconstruction and the application of appropriate corrective techniques seems to improve early surgical results, and extend the indications for this procedure whilst maintaining the already reported good mid-term results.
Appendix A Conference discussion
Dr R. Dion (Leiden, The Netherlands): You have described three methods of repair: plication, resection and free edge reinforcement. When exactly do you use the 7/0 Gore-Tex stitch to cure a cusp prolapse?
Dr El Khoury: The Gore-Tex is used in two situations. When the prolapse is corrected by plication or resection and the commissural areas are fragilized or the free margin is fragilized, I will add only this Gore-Tex to reinforce the repair and to stabilize the repair. When the repair is not corrected by plication or resection, sometimes I could use this free margin reinforcement to correct the prolapse by exerting some gradual tension on the over and over running suture, and so the level of coaptation is brought up. So free edge reinforcement is used to reinforce the free margin or to correct the prolapse.
Dr A. Moritz (Frankfurt, Germany): We also have a renewed interest in aortic valve repair. What else other than eyeballing is a method you have to judge leaflet prolapse, one?
And second, usually running sutures tend to stretch over time. We know this from aortotomy closure or we have not so good an experience with the running suture on mitral valve restriction or annular ring. So I am not sure how long this type of suture repair of the leaflet edge will stay.
Dr El Khoury: For the first question, the most important thing is really to expose very well the aortic root and the valve by a transverse aortotomy, putting three stitches on each commissure, doing a vertical traction on each commissure and have a look on the leaflet. A prolapse, an absolute prolapse, or let's say a real prolapse, when your commissures are really pulled up is when your free margin is below the aortic annulus.
Now, if you put radial tension on the commissure with a central stitch in the middle of each leaflet, you could compare the lengths of the normal, let's say, and the prolapsed leaflet, and it could give you an idea about the difference in length between the prolapsing and nonprolapsing leaflet. I think playing with the commissures and the central stitch could give you a clear idea about the presence of the prolapse and the amount of tissue to be repaired.
Now, for the second question, I think it is possible that this running stitch will be stretched or something like that, but I think, as for the mitral valve, if we could obtain immediately post-operatively very nice coaptation, the stress on the free margin will be lower and maybe the stress on the running suture is not too important to change the length of the free margin.
