| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
a Division of Pediatric Cardiovascular Surgery, The Heart Institute for Children at Advocate Hope Children's Hospital, Oak Lawn, Illinois
b Division of Pediatric Cardiovascular Surgery and Thoracic Surgery, Rush University Medical Center, Chicago, Illinois
Accepted for publication May 17, 2010.
* Address correspondence to Dr Polimenakos, Department of Surgery, The Heart Institute for Children, Hope Children's Hospital, Congenital and Pediatric Cardiovascular Surgery, Rush University Medical College, 1653 W Congress Pkwy, Chicago, IL 60612-3244 (Email: anastasios_c_polimenakos{at}rush.edu).
Presented at the Forty-sixth Annual Meeting of The Society of Thoracic Surgeons, Fort Lauderdale, FL, Jan 25–27, 2010.
 |
Abstract |
|---|
 Top Abstract IntroductionMaterial and MethodsResultsCommentDiscussionReferences |
|---|
Methods: From July 1987 to December 2008, 78 patients younger than 10 years of age underwent aortic cusp extension valvuloplasty in the form of pericardial cusp extension and selective use of tricuspidization. Sixteen (20.5%) patients were younger than 1 year of age. Twenty-seven had bicuspid aortic valve, 34, congenital aortic valve stenosis, and 17, congenital or acquired aortic insufficiency or aortic stenosis. Forty-two patients had balloon valvuloplasty or surgical valvotomy before aortic cusp extension valvuloplasty. Median follow-up was 12.4 years (range, 0.1 to 21.6 years). Freedom from aortic valve replacement (AVR) and determinants of outcome were analyzed.
Results: There were no early or late deaths. During the follow-up period, 23 patients (29.5%) had Ross operation and 8 patients (10.2%) had other AVR. The z values of left ventricular end-diastolic dimension, aortic annulus, aortic sinus diameter, sinotubular junction diameter, and left ventricular wall thickness before AVR were 3.8 ± 2.95, 2.1 ± 1.15, 4.2 ± 1.22, 1.78 ± 1.24, and 2.92 ± 1.31, respectively. Actuarial freedom from AVR at 1, 5, and 10 years was 97.3 ± 2.0%, 71.3 ± 5.8%, and 55.6 ± 6.9%, respectively.
Conclusions: Aortic cusp extension valvuloplasty with tricuspidization allows left ventricular reverse remodeling with satisfactory long-term durability and freedom from AVR. Used selectively, it represents a reliable and effective approach in infants and children with congenital or acquired abnormal aortic valve.
|
Introduction |
|---|
|
TopAbstractIntroductionMaterial and MethodsResultsCommentDiscussionReferences |
|---|
Growth potential and availability of pulmonary autograft are major advantages of the Ross procedure, especially in the pediatric population. Despite being technically demanding, ACEV may be particularly advantageous in infants and children relative to other alternatives, including freedom from reoperation and need for anticoagulation when mechanical prostheses are used, or complications inherent to the Ross procedure.
Such repair techniques in infants and small children, mostly with limited follow-up, have been published [6–11]. Outcome analysis after ACEV with selective use of tricuspidization was undertaken. Freedom from AVR and determinants of outcome were assessed.
|
Material and Methods |
|---|
|
TopAbstractIntroductionMaterial and MethodsResultsCommentDiscussionReferences |
|---|
|
Methods
A retrospective analysis of preoperative and postoperative data was conducted.
Echocardiographic evaluation was performed on all patients preoperatively, intraoperatively, and at regular intervals (6 to 12 months) postoperatively. Echocardiograms were reviewed by an independent echocardiographer who was blinded to the study.
Grading of AI was performed as (1) none or trace, (2) mild (no LV dilation, no retrograde flow in the descending aorta, AI vena contracta width <4 mm), (3) moderate (LV end-diastolic dimension [LVEDD] z value >2 and <4, retrograde flow in the descending aorta, vena contracta 4 to 6 mm), or (4) severe (LVEDD z value >4, retrograde flow in the descending aorta, vena contracta >6 mm). Aortic stenosis was estimated by Doppler study, and the peak instantaneous gradient was calculated. Peak instantaneous gradient less than 40 mm Hg was considered mild, 41 to 75 mm Hg moderate, and 76 mm Hg or greater severe. Aortic valve annulus and root diameters (maximum dimension at the level of the sinuses) were measured from parasternal long-axis views.
Left ventricular dimensions, volumes, and mass were measured from apical four-chamber and cross-sectional views. Short-axis LVEDD was measured. Data were indexed to body surface area and reported as z values.
Patients were not randomized. Each case was managed according to certain anatomic and functional valve criteria. Patients selected for ACEV had an aortic annulus z value of –1.5 or greater without LV or mitral annulus hypoplasia, and several echocardiographic or Doppler-derived indices that determine severity of valve dysfunction and timing for ACEV. These include (1) regurgitant jet-to-annulus diameter ratio of 35% or greater or progressive increase LVEDD z value of +2.5 or greater, (2) peak instantaneous gradient of 40 mm Hg or greater associated with progressive LV hypertrophy, or (3) mixed lesions with a variable degree of AI and AS.
Tricuspidization was added in BAV with eccentric opening, in cases in which raphe were well developed, and patients with aortic valve of limited cusp mobility away from the hinge point.
Intraoperatively, compromised mobility of cusp at the hinge point, extensive and multiple cusp dysplasia, and extensive commissural fibrosis extending to the coronary ostia were relative contraindications.
Operative Technique
The basic concept of ACEV was previously described [12]. After median sternotomy a piece of autologous pericardium is harvested, carefully trimmed and thinned from fat, and treated with a 0.625% glutaraldehyde solution. Early in our experience the pericardium was treated with glutaraldehyde solution for 10 minutes. As our technique evolved, 3-minute treatment under aseptic conditions was used and the pericardium kept moist with normal saline solution. Aortobicaval cardiopulmonary bypass with moderate hypothermia (32°C), antegrade and retrograde myocardial preservation is used. After an oblique aortotomy incision each cusp is evaluated as to the extent of tissue deficiency, the shape, and the irregularities of the free edge. Only each cusp's thickened free margin and body are thinned out, leaving its base and unaffected body intact. When tricuspidization is needed, the fused cusp is cut at the raphe precisely to the aortic wall. Any subcommissural fused tissue is released. A pericardial extension is then fashioned to fit the specific architecture of each cusp, but slightly oversized in depth (10% to 15%) and length (up to 25%). Continuous 5-0 or 6-0 polypropylene suture is used. The sutures are placed from the cusp's center toward each commissure. The suture line on the pericardial site is slightly wider than that on the cusp to support a generous mural edge. The depth of each sinus is assessed. Each neocusp's free edge is leveled with the sinotubular bar at the commissural level but more caudally at the center. The commissural ends are suspended at the level of the sinotubular bar using transmural pledgeted polypropylene sutures. Suspension is tailored to provide optimal coaptation, avoid crowding of the subcommissural triangle, and reestablish normal semilunar appearance of each neocusp. When severe dilatation of the ventricular–aortic junction is present, a reduction annuloplasty at the subcommissural area between right and left cusps is performed. When the cusp is prolapsed, no attempt is made to excise any portions, but the pericardial extension is sutured to the cusp's free margin and, consequently, suspended to the aortic wall as described (Fig 1).
|
Statistical Analysis
Data are expressed as mean ± standard deviation for continuous variables and as median and range for categorical variables. For comparison of continuous or categorical variables between groups, independent samples Student's t test and 
2 analyses, respectively, were used. Fisher's exact test was used to detect significant differences between groups. Unpaired two-tailed Student's t test was performed for comparison of continuous variables between groups. The probability of freedom from events was estimated according to the Kaplan-Meier method, and estimates were then compared with the log-rank test. Univariate analysis was carried out using a probability value of less than 0.05 to determine predictors of freedom from AVR. Significant factors were entered into the multivariate logistic regression model to assess their independent impact. Owing to high degree of covariance of variables, stepwise analysis was performed repeatedly to identify the most important variables. SPSS 15.0.1 for Windows (SPSS Inc, Chicago, IL) was used.
|
Results |
|---|
|
TopAbstractIntroductionMaterial and MethodsResultsCommentDiscussionReferences |
|---|
Seventeen patients underwent at least one intervention before ACEV in our institute. Eighteen BV were performed in 13 patients. Four underwent SV. Another 25 patients (9 with BAV and 16 with congenital AS) had BV or SV in outside facilities.
Before ACEV, AI was severe in 29 patients (37.2%) and moderate in 6 patients (7.7%). Fourteen with severe AI had a tricommissural valve, 1 a unicommissural valve, and the remaining bicommissural valves. Among patients with moderate AI, the condition occurred after arterial switch operation in 1, after BV in 4, and after surgical repair for aortic endocarditis and root abscess in 1. Combination of predominantly moderate AI (with variable degree of AS) was present in 23 patients (29.5%). Moderate AS was present in 20 patients (25.6%); 14 after BV or SV for congenital AS, 5 as a result of BAV, and 1 after ventricular septal defect repair. Diagnoses and additional procedures are summarized in Table 1.
Mean age and weight at ACEV were 6.9 ± 2.3 years and 16.1 ± 5.71 kg, respectively. Sixteen (20.5%) patients were younger than 1 year old at ACEV. Cross-clamp and cardiopulmonary bypass times were 74.7 ± 23.2 minutes and 105.1 ± 32.7 minutes, respectively. Intraoperatively, reinstitution of cardiopulmonary bypass was necessary in 5 (6.6%) patients for residual moderate or greater AS, AI, or both after ACEV. A second ACEV was successfully performed in 4 patients (with only trace or mild AI). One had AVR.
There was no hospital death. One patient had postoperative sinoatrial node dysfunction. One experienced a third-degree block requiring a pacemaker before hospital discharge (Table 2). At hospital discharge AI or AS was improved in 77 patients (Table 3).
|
|
|
When BV or SV was used before ACEV, AVR was performed at a mean time of 5.32 ± 1.95 years (versus 5.52 ± 2.23 years when no BV or SV was used; p = 0.14). Seven patients who had ACEV during infancy needed AVR at a mean time of 6.86 ± 2.82 years. Twenty-three (of 31 AVR) patients underwent Ross operation. Konno was added in 2 Ross cases. Eight (10.2%) had AVR with a mechanical valve (n = 5) or a bioprosthesis (n = 3; Table 2). Valve disease at the time of AVR is shown in Table 5.
|
|
|
|
Comment |
|---|
|
TopAbstractIntroductionMaterial and MethodsResultsCommentDiscussionReferences |
|---|
Patients with complex anatomy, in whom ACEV was exclusively used, were included in the analysis. Lesions (isolated commissural fusion or single cusp prolapse or tear), for which standard non-ACEV techniques have proved durable [8, 13], were excluded. In patients with severe or critical AS, BV or SV can be an effective initial strategy [14, 15].
Aortic cusp extension valvuloplasty, when tricuspidization is selectively added (as in this series), effectively reduced AI and AS and improved LV wall thickness and LV dimensions in infants and children. Promising early outcomes have been reported in adolescents with congenital or acquired aortic valve disease. Most series, though, are relatively small, or include a combination of simple and complex repairs [8–11, 16–19]. As shown, this repair strategy allows expeditious LV remodeling even in patients with significant LV dilation or LV wall thickness with satisfactory long-term durability and freedom from AVR. The z values of LVEDD, aortic annulus, aortic sinus diameter, sinotubular junction diameter, and LV wall thickness, which improved after ACEV, remained relatively stable in patients who met no AVR criteria at last echocardiographic follow-up after ACEV.
Use of a cusp extension as part of an aortic valve repair strategy in patients with rheumatic AI and a structurally normal aortic valve can be fairly durable [16, 17]. Our experience and that of others [12, 20, 21] indicate that ACEV provides a comprehensive repair and restores all the important anatomic features of the aortic valve. Most critical among these, in a congenitally abnormal valve, are the attenuated subcommissural triangles, the foreshortening of the cusp's free margin, the shallow sinus(es) of Valsalva, and the eccentric and uneven orifice opening. Cusp extensions counteract the valve's inherent sinus(es) of Valsalva shallowness, reestablish normal depth of the sinuses, secure adequate and longer coaptation surface, and restore the normal "crownlike" appearance of the valve. Cusp resuspension at the level of the sinotubular bar, aiming at a wider subcommissural triangle, allows more freedom of the cusp's movement. Tricuspidization ensures a larger central opening and minimizes turbulence. Several substitutes have been used to augment the cusps. These include autologous or bovine pericardium, and polytetrafluoroethylene membrane. Satisfactory outcomes in the pulmonary position have been reported with thin polytetrafluoroethylene membrane. Glutaraldehyde-treated autologous pericardium has been most consistently used in the aortic position. Although glutaraldehyde promotes calcification and fibrosis, it is needed for collagen fiber linkage and, hence, tissue strength. The concentration and duration of treatment with glutaraldehyde did change during the study period. Currently, a 0.625% glutaraldehyde solution for a shorter duration (3 minutes) is used, compared with our early experience [22]. This might promote increased pericardial pliability and minimizes the incidence of excessive stiffness and early calcification of the reconstructed cusps. Our sample, though, is too small to draw valid inferences as to what constitutes the optimal concentration and duration of treatment. An ongoing study in our institute will attempt to assess whether concentration and duration of treatment with glutaraldehyde influence early and long-term outcomes after ACEV.
Indications for surgery in aortic valve disease are often based on experience with adult patients. In a pediatric patient any delay may be detrimental because of the longer life span. In addition, AI after BV can be particularly deleterious during infancy owing to its disruptive nature on both valvar planes and cusp integrity [9, 14, 15]. The encouraging short- and long-term results of cusp extension valvuloplasty techniques may sway the timing for intervention toward earlier surgery [9–12, 18–21].
Published studies have demonstrated progressively decreasing freedom from aortic valve reintervention after various cusp extension valvuloplasty techniques [9–12, 16–21]. Ours is among the largest series for patients younger than 10 years of age with satisfactory long-term freedom from AVR. As expected, moderate or greater recurrent AI or AS at interval follow-up after ACEV was an independent predictor of failure. The significance of the additional procedure during ACEV, as a determinant of outcome, was a surprising finding. It might indicate the impact of subvalvar or supravalvar repair (4 of 6 additional procedures) on the long-term durability of ACEV. With respect to freedom from AI or AS after ACEV, outcomes are often unpredictable. Some patients exhibit restenosis, whereas others have recurrent AI or both. The progressive AS observed in some of our patients during follow-up may be related to fibrosis and calcification of the pericardial extensions. It has been observed earlier in our experience when the cusp augmentation was generous and ACEV was performed on a few patients with a valve annulus z value less than 1.5. The Ross or Ross-Konno procedure is the preferred choice for these patients.
Despite satisfactory long-term outcomes [3–5, 23–25], the Ross procedure requires valved-conduit replacement in the right ventricular outflow and is associated with increasing evidence of progressive neoaortic root dilation or autograft failure. Small valve prostheses, especially in infants and children, can be associated with patient–prosthesis mismatch as the child outgrows the valve, morbidity related to long-term anticoagulation, and the need for future valve replacement [5, 23–25]. Aortic cusp extension valvuloplasty with selective tricuspidization provides an alternative that can neutralize many of the drawbacks of early valve replacement. It has extremely low operative early or long-term mortality, can be associated with reproducible results, and arrests the disease process until such an age that valve replacement becomes more advantageous. As a result, with improved patient selection and refinement of cusp extension valvuloplasty techniques, the durability of ACEV should continue to improve.
As with any retrospective study, there are inherent limitations that may have introduced bias. This experience encompasses a period when the technical details of ACEV were under development. It cannot be considered for direct comparison between ACEV and AVR. Considering that our institute is a referral center for aortic cusp extension reconstructive surgery, this study might represent a more severe expression of this population. Finally, serial echocardiographic assessment of pre-ACEV aortic valve and LV was not available, which limits our insight into the evolution of changes in aortic valve and LV morphology, function, and flow kinetics with time.
In conclusion, like others [10, 11, 18, 20, 21, 26] our results with ACEV in infants and children with complex congenital or acquired aortic valve disease encourage earlier and more aggressive management considering that LV has to preserve function for a longer life span. The aortic annulus z value, transvalvular pressure gradient, LV hypertrophy progression, LV dimensions (such as LVEDD z value), and regurgitant jet-to-annulus diameter ratio may guide optimal timing and strategy. When certain anatomic and functional valve criteria are met, ACEV with selective use of tricuspidization is a safe, effective, and reproducible surgical choice. It allows expeditious LV reverse remodeling with satisfactory long-term durability and freedom from AVR. Multiinstitutional studies comparing such repair techniques with AVR are needed to refine selection criteria and determine patients' suitability for each strategy.
|
Discussion |
|---|
|
TopAbstractIntroductionMaterial and MethodsResultsCommentDiscussionReferences |
|---|
I would like to compliment Dr Polimenakos and Dr Ilbawi and the other authors for their superb results in a large series of 78 patients under 10 years of age, with 16 under 1 year of age. They had no early and no late deaths, which is quite a superb outcome. They also are very proficient in the technique, which has to be learned, and by 6 months only 1 patient had moderate aortic valve regurgitation.
I have no disclosures.
We began our experience with pericardial leaflet extension in the '90s and gave our first reports in 2003, '05 and '07. We included in this series 4 patients, neonates with truncus arteriosus, who had both truncal valve regurgitation and stenosis, and found this procedure extremely effective in successfully getting a competent valve and relieving the stenosis, even in the neonate, and these patients had redo operations about a year and a half later where, for the second time, pericardial leaflets were used for the second repair.
Our technique is very similar to that which was described by the authors, and the outcomes were similar, with redo operations being required at an average of about 4 to 4.5 years after the first procedure, much more early in those who are very young, such as neonates or small infants, and the older children can go as long as 7 or 8 years, and adolescents as long as 10 years.
With our technique, we have also found that the bicuspid aortic valve, whether it is stenotic or regurgitant, should be tricuspidalized and not maintained as a bicuspid valve, and we have found greater durability and also better competence with a three-leaflet valve. Each leaflet has to be carefully measured to equal the diameter of the aorta, and we extended the leaflets, as the authors did, up to just below the sinotubular junction.
We began this, and a great tribute needs to be given to Carlos Duran, who first described the use of the pericardium in this fashion, and also instituted the short duration of fixation of only 9 minutes. We began with 9 minutes, and in our younger patients with smaller leaflets, we went on to use only 3 minutes, first 6 and then 3, similar to the authors, for the very small infants.
We reported on a pathologic study, and, to our surprise, we found that in the bicuspid aortic valves there was more calcification in the native remnant of the leaflet than in the glutaraldehyde-treated pericardium. However, the glutaraldehyde pericardium was embedded in a relatively thick layer of fibrous tissue on both sides, which, although pliable, still made it stiffer than we would have liked.
My questions for the authors are: What are your current indications for the Ross procedure? Have you had experience with redo pericardial leaflet extension? Have you noticed any decrease in the thickness and fibrosis with a 3-minute fixation period? And what are your thoughts about the use of substitutes like CorMatrix, particularly for patients who have had previous median sternotomies where the pericardium becomes very roughened and thickened because of the redo?
Thank you very much.
DR POLIMENAKOS: These are all great questions. With regards to the first question, we do use the Ross operation. Actually early in our series, we used the valvuloplasty technique, even with more severe forms of aortic stenosis. We found that probably secondary to two major factors such as (1) extent of the time and duration of preparation with 0.625% of glutaraldehyde of the pericardial extension (up to 15 minutes earlier in the series), and (2) z value of the aortic valve annulus and the performance of the valve, which can play a significant role in the recurrence of predominantly AS, this subset of patients with more than moderate pre-ACEV AS were associated with substantially higher recurrence of AS after ACEV. That is why in this particular subset of patients we would most likely recommend the Ross procedure, or Ross-Konno, of course, if they meet the criteria for a Konno operation.
With regards to the second question, at the present time we do not advocate a second redo reconstructive approach. We had only 8 of these 12 patients that we had to go back and redo the ACEV because in the original series we did not use the tricuspidization technique as part of the procedure. And I totally agree with you, the performance of the valve is superb, particularly the first 3 years after tricuspidization, especially with a bicuspid aortic valve.
With regards to intraoperative findings, we found, as yourself, some fibrosis at the level of the reconstructed neocusp, as well as accelerated calcification at the level of the commissures, especially in the bicuspid aortic valves.
Finally, we started using CorMatrix in reconstructive approaches for the right-sided valves. We haven't tested that, and we haven't tested any other materials, including PTFE (polytetrafluoroethylene) for the aortic valvuloplasty. So I am not sure whether or not these materials can perform the same way that glutaraldehyde-treated autologous pericardium performs. In addition, we haven't yet restudied those patients where, since 2004, autologous pericardial extensions processed with glutaraldehyde for only 3 minutes was used. So we don't know whether or not fibrosis and calcification remains an issue.
Thank you.
|
References |
|---|
|
TopAbstractIntroductionMaterial and MethodsResultsCommentDiscussionReferences |
|---|
This article has been cited by other articles:
|
|
 |
|
  M. Nosal', R. Poruban, P. Valentik, M. Sagat, A. S. Nagi, and A. Kantorova Initial experience with polytetrafluoroethylene leaflet extensions for aortic valve repair Eur J Cardiothorac Surg, June 1, 2012; 41(6): 1255 - 1258. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 Y. d'Udekem Aortic valve surgery in children Heart, July 15, 2011; 97(14): 1182 - 1189. [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| ANN THORAC SURG | ASIAN CARDIOVASC THORAC ANN | EUR J CARDIOTHORAC SURG |
| J THORAC CARDIOVASC SURG | ICVTS | ALL CTSNet JOURNALS |
