| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Ann Thorac Surg 1999;68:1350-1355
© 1999 The Society of Thoracic Surgeons
a Department of Cardiovascular Surgery, Children’s Hospital of Michigan, Wayne State University School of Medicine, Detroit, Michigan, USA
Address reprint requests to Dr Walters, Department of Cardiovascular Surgery, Children’s Hospital of Michigan, 3901 Beaubien Blvd, Detroit, MI 48201
e-mail: hwalters{at}dmc.org
Presented at the Thirty-fifth Annual Meeting of The Society of Thoracic Surgeons, San Antonio, TX, Jan 25–27, 1999.
 |
Abstract |
|---|
 Top Abstract IntroductionPatients and methodsResultsCommentReferences |
|---|
Methods. A univariate analysis was performed in this retrospective study to identify the possible risk factors for failure of the repair resulting in the need for AV replacement (AVR).
Results. The study included 24 patients, 15 (62%) boys and 9 (38%) girls, with a mean age of 9.1 ± 1.2 (SEM) years. The VSD was perimembranous in 15 (62%) and subarterial in 9 (38%). The prolapsed aortic cusp was the right in 13 (54%), the noncoronary in 6 (25%), and both in 5 (21%). Plication was performed at one end of the free edge of the prolapsed cusp(s) in 12 (50%) and at more than one end in 12 (50%) of the patients. The VSD was closed by use of a patch in 21 (88%) and by direct suture closure in 3 (12%). At the mean follow up of 7.3 ± 1.3 years, the degree of AI was none in 6 (25%), trivial in 5 (21%), mild in 9 (38%), moderate in 1 (4%), and severe in 3 (12%). The 15-year actuarial freedom from reoperation was 81% ± 19% (95% confidence limit). By univariate analysis, the possible risk factors for AV repair failure were the degree of AI at hospital discharge (p = 0.004), direct closure of the VSD (p = 0.061), smaller size of the VSD (p = 0.081), and plication of more than one end of the prolapsed cusp(s) (p = 0.095).
Conclusions. Trusler’s AV repair is an effective and durable technique for the surgical treatment of patients with VSD-AI syndrome. The adequacy of the initial repair is the most important determinant of the long-term results.
|
Introduction |
|---|
|
TopAbstractIntroductionPatients and methodsResultsCommentReferences |
|---|
|
Patients and methods |
|---|
|
TopAbstractIntroductionPatients and methodsResultsCommentReferences |
|---|
Operative technique
In all patients, the operation was performed through a median sternotomy incision. Cardiopulmonary bypass was established using bicaval cannulation, a left ventricular vent, and moderate hypothermia. The heart was arrested with the antegrade administration of dilute, sanguineous cardioplegia delivered into the aortic root while supporting the aortic leaflets using external digital compression of the aortic root. The aorta was opened through a curvilinear incision into the noncoronary sinus. Additional cardioplegia was delivered directly into the coronary ostia to supplement the initial arresting dose. Retrograde cardioplegia was occasionally used. The morphology of the aortic valve cusps was then examined carefully with special attention to the identification of elongation and redundancy with prolapse. A 6-0 Prolene Frater’s stitch [3] was then placed in the corpora arantii of the three aortic cusps. This facilitated more precise identification of the redundant free edge(s) of the prolapsed cusp(s). The excess length of the leaflet edge was then folded at the commissure. This fold of the prolapsed cusp was then secured against the aortic wall using a pledgeted 4-0 or 5-0 Prolene horizontal mattress suture, with one pledget placed on the inside of the aorta and another placed on the outside. The Frater’s stitch was then removed. Aortic valve competence was tested by the injection of cold saline into the aortic root. The aortotomy was then closed. The adequacy of the repair was then tested again by injecting antegrade cardioplegia and observing satisfactory aortic root distention. The degree of residual AI could, at times, also be assessed during antegrade cardioplegia administration by observing the aortic valve directly through the VSD. The VSD was closed either through the right atrium (n = 2, 8%) or the right ventricle (n = 22, 92%).
Data collection and statistical analysis
All data were sorted on the Microsoft Excel for Windows program version 7.0 (Microsoft Corp, Redmond, WA) and were exported into SPSS for Windows release 8.0 (SPSS Inc, Chicago, IL) for coding and statistical analysis. Quantitative variables that approximated a normal distribution were reported as the mean plus or minus the standard error of the mean, and were analyzed by the Student’s unpaired t test. Quantitative variables that did not approximate a normal distribution were reported as the mean plus or minus the standard error of the mean, with the median, and were analyzed by the Mann-Whitney test (Wilcoxon rank-sum test). Nominal variables were analyzed nonparametrically by the Fisher’s exact or the 
2 test. A p value less than 0.05 was selected to define statistical significance. A p value between 0.05 and 0.10 was considered suggestive. Because of the small size of the population, a multivariate analysis was not performed.
|
Results |
|---|
|
TopAbstractIntroductionPatients and methodsResultsCommentReferences |
|---|
Morphology
Fifteen (62%) VSDs were perimembranous and nine (38%) were subarterial (type 1, supracristal, doubly committed/juxta-arterial). The size of the VSD was small in 4 (17%), moderate in 11 (46%), and large in 9 (37%) of the patients. Neither the PGLV-RV or the PRV/PLV correlated with the size of the VSD observed in the operating room. The prolapsing coronary cusp was the right in 13 (54%), the noncoronary in 6 (25%), and both the right and noncoronary in 5 (21%) of the patients.
Operative procedure
Plication of the free edge of the prolapsed cusp at one end was performed in 12 (50%) patients. More than one plication had to be performed in the remaining 12 (50%) patients. The VSD was closed using a patch in 21 (88%) and by primary suture closure in 3 (12%) of the patients. Four (17%) patients had associated cardiac procedures in the form of resection of a subaortic membrane in 2 and closure of a fenestration in the right coronary cusp in another 2 patients.
Postoperative results
There was no hospital mortality and there have been no late deaths to date. The degree of AI at the patient’s discharge, as assessed by both clinical and by echocardiographic examination, was none in 5 (21%), trivial in 6 (25%), mild in 10 (42%), moderate in 2 (8%), and severe in 1 (4%). At the most recent follow-up, the degree of AI, as assessed by echocardiography, was none in 6 (25%), trivial in 5 (21%), mild in 9 (38%), moderate in 1 (4%), and severe in 3 (12%). All 3 patients with severe AI underwent aortic valve replacement. The remaining patient with moderate AI is now followed at frequent intervals to assess the significance of the AI. The 15-year actuarial freedom from aortic valve replacement for failure of the repair was 81% ± 19% (95% CL) (Fig 1). The long-term adequacy of the repair is depicted in Figure 2.
|
|
|
|
Comment |
|---|
|
TopAbstractIntroductionPatients and methodsResultsCommentReferences |
|---|
Pathogenesis
The mechanism of aortic valve cusp prolapse is not known. Van Praagh and McNamara [1] assert that three anatomic factors are responsible for normal aortic valve competence: commissural support from above, leaflet support by diastolic apposition, and infundibular (conal) septal support from below. Aortic insufficiency in VSD-AI syndrome can be partially explained by a lack of infundibular (conal) septal support from below. It is unlikely that this is the entire explanation because most large perimembranous defects are closely related to the aortic valve and very few are associated with aortic valve prolapse with or without incompetence. Quite likely, there is an intrinsic structural abnormality of the aortic valve that, in addition to the VSD, predisposes one or more cusps to progressively prolapse. Yacoub and associates postulated that this basic structural abnormality is a progressive discontinuity between the aortic valve annulus and the aortic media [6]. Hemodynamic factors also probably aggravate the tendencies toward the development of aortic insufficiency. These effects are present throughout the entire cardiac cycle. A Venturi effect is created by the left-to-right shunting through the VSD during early systole. This turbulent flow tends to displace the aortic cusp into the right ventricular cavity [14]. Later, in systole, the aortic cusp is pushed further into the right ventricular cavity by direct left ventricular pressure. In diastole, the aortic root pressure pushes the aortic cusp further downward. This results in a failure of proper coaptation of the aortic valve cusps and produces progressive aortic insufficiency.
Timing of the operation
We advocate closure of VSDs, regardless of the type, the size, or the apparent absence of aortic valve prolapse, when any degree of AI is identified. We also close all subarterial VSDs associated with aortic valve prolapse, even in the absence of AI, because the likelihood of spontaneous VSD closure is low. For asymptomatic, perimembranous VSDs in patients with aortic valve prolapse but without AI or other classical indications for VSD closure, the decision for surgical repair is more controversial; however, we tend to repair these patients also. Certainly, these patients must undergo VSD closure if any degree of AI develops or if they develop any other criteria for VSD closure such as symptoms, enlarging left ventricular end-diastolic dimensions, or evidence of significant pulmonary hypertension. The apparent echocardiographic or angiographic size of the VSD, the PRV/PLV ratio, and the PGLV-RV are not reliable indicators of the true size of the VSD in the presence of aortic valve prolapse. Thus, the true size of the VSD tends to be underestimated, perhaps leading to a false sense of confidence that the VSD is anatomically insignificant and is likely to close spontaneously. The presence of aortic valve prolapse in the absence of AI should alert the clinician to the existence of important hemodynamic forces that, in concert with the underlying VSD, constitutes a significant clinical condition. Surgical repair should be strongly considered.
Surgical repair techniques
Because the AI in VSD-AI syndrome is associated with prolapse of the aortic valve cusp(s) with elongation of the free edge, initial attempts to surgically treat AI were directed toward aortic valve repair. Starr and colleagues [15] were the first, in 1960, to successfully apply aortic valvuloplasty techniques to repair of the VSD-AI syndrome by shortening the free edge of the prolapsed cusp. Spencer and associates [16], in 1962, produced satisfactory aortic valve competence in 2 patients by plicating the free margin of the prolapsing cusp. Plauth and colleagues [2] stressed the importance of accurate measurement of the free edge of the prolapsed cusp, and used an approximating suture in the corpora arantii to make the free margin of prolapsed cusp exactly equal to that of the adjacent normal cusp by plication. Frater [3], in 1967, described a simple technique for comparing the lengths of the free edge of the aortic cusps by suspending the three corpora arantii using a single 6-0 Prolene suture. During this same era, some authors reported unsatisfactory results using aortic valvuloplasty techniques [2,17–19]. These results stimulated a movement toward prosthetic aortic valve replacement rather than repair for the surgical treatment of patients with VSD-AI syndrome [18]. In 1973, Spencer and associates [4] rekindled optimism for aortic valvuloplasty by presenting their successful experience with 20 patients from three different centers. Later in the same year, Trusler and associates [7] reported their experience in 16 children using a technique that combined Frater’s stitch with plication of the elongated portion(s) of the aortic cusp(s) against the aortic wall. Trusler and associates stabilized this repair by reinforcing the adjacent commissure(s) with a hood of fine Dacron. Spencer and associates, in discussing Trusler’s paper, noted that previously reported repair techniques that excluded securing the plicated cusp to the aortic wall and commissural reinforcement produced a more mobile repair site but probably resulted in more repair failures due to instability [20].
In 1990, Carpentier’s group described yet another technique of aortic valve repair using triangular resection of the prolapsed cusp, aortic annuloplasty, and transaortic patch closure of the VSD [5]. Yacoub and associates [6], in 1997, described an aortic valvuloplasty technique based upon their anatomic observations of the VSD/AI syndrome. This technique includes transaortic primary closure of the VSD using a series of pericardial pledgeted mattress sutures inserted into the right ventricular side of the crest of the interventricular septum. These same sutures are passed through the aortic valve annulus and then through the thin portion of the prolapsed aortic sinus. When these sutures are secured, the result is primary closure of the VSD, elevation of the aortic valve annulus, reduction in the size of the prolapsed and dilated aortic valve sinus, and a reduction in aortic insufficiency.
When the VSD is associated with trivial AI, we perform patch closure of the VSD without exploring the aortic valve. When the AI is mild, exploration of the aortic valve is sometimes undertaken if preoperative echocardiographic findings suggest significant aortic valve prolapse or if other parameters, such as wide pulse pressure or low aortic root pressure with cardioplegic administration, suggest that the degree of preoperative AI was underestimated. All patients with moderate or severe AI undergo aortic valvuloplasty in addition to VSD closure.
We now recommend patch closure of all VSDs, regardless of size or other anatomic features, because of our finding of a possible association of primary VSD closure with failure of the aortic valvuloplasty. Others [4, 13] have also reported that primary VSD closure may be a risk factor for valvuloplasty failure. It is possible that primary VSD closure distorts the aortic valve annulus and, thereby, compromises the aortic valve repair. Conversely, this finding may not relate to the repair technique at all. It is possible that smaller VSDs, those that are most likely to be repaired primarily, are associated with a greater Venturi effect. This may tend to produce earlier and greater damage to the aortic valve cusps and predispose to repair failure. We are reluctant to endorse Yacoub’s surgical technique of primary VSD closure in VSD-AI syndrome [6], due to concerns that any undue tension created by the technique might result in avoidable VSD recurrences and/or conduction disturbances. Most of the patients in our series were operated upon during an earlier era when a right ventriculotomy was our preferred approach to the VSD. In the present era, we prefer to close the VSD through the right atrium or the pulmonary valve.
Of the various repair techniques described above, we have consistently employed a modification of the Trussler repair in patients with VSD-AI syndrome. The excess length of the prolapsed aortic valve cusp is plicated against the aortic wall using a horizontal mattress suture reinforced with a Teflon or pericardial pledget. We do not routinely place a hood over the adjacent commissure because of concerns about reducing the cross-sectional dimensions of the aortic valve.
Because the adequacy of the initial repair was the most important determinant of long-term results in our study, we recommend close preoperative observation and early operative intervention before excessive elongation and prolapse of the aortic valve cusp(s) is allowed to occur. This recommendation is further supported by our finding that valvuloplasty failure may be related to the number of plication sutures needed to achieve a satisfactory repair. Okita and colleagues, in their multivariate analysis, confirmed that the number of plication sutures represent an independent risk factor for valve repair failure [8]. The routine use of intraoperative transesophageal echocardiography provides an immediate assessment of the adequacy of the aortic valvuloplasty and may improve the long-term results.
Morbidity and mortality
The morbidity and mortality associated with VSD closure combined with aortic valvuloplasty is low. In our series, there were no hospital or late postoperative deaths. Okita and colleagues reported a hospital mortality of 1.6% with no late deaths, resulting in a 15-year actuarial survival of 98.3% [8]. Trusler and associates experienced no hospital deaths with two late deaths, yielding a 10-year actuarial survival of 96% [9].
Freedom from reoperation
Our 15-year actuarial freedom from repair failure of 81% ± 19% (95% CL) compares favorably with other published series, where estimates range from 85% at 10 years [9] to 64% at 15 years [8]. Rhodes and colleagues reported an 18-year freedom from reoperation of 51% [13].
There were three repair failures necessitating aortic valve replacement. All 3 of these patients had moderate to severe AI at the time of their hospital discharge. None of the patients who were discharged with absent, trivial, or mild AI required reoperation. Moreover, the durability of the repair in this latter group of patients has been superb. All of them have retained the same degree of postoperative AI except for 1 asymptomatic patient who has progressed to moderate AI. This finding supports the contention that VSD closure with a modified Trusler’s repair of the aortic valve is not only effective in achieving an immediate reduction in the degree of preoperative AI, but also stabilizes the valve pathology to yield excellent long-term aortic valve competence.
Criticism of the study
The limitations of this study relate to the inherent liabilities of all retrospective analyses. Also, the small number of patients precluded the performance of a multivariate analysis to identify independent risk factors for valve repair failure.
Inferences
Trusler’s aortic valve repair is an effective and durable technique for the surgical treatment of patients with VSD-AI syndrome and is associated with a low morbidity and mortality. The adequacy of the initial repair is the most important determinant of the long-term results. This underscores the importance of early VSD closure with and without aortic valve repair before the development of excessive cusp elongation and prolapse. We recommend that all VSDs associated with AI, regardless of size, should be closed with a patch.
|
References |
|---|
|
TopAbstractIntroductionPatients and methodsResultsCommentReferences |
|---|
This article has been cited by other articles:
|
|
 |
|
  S.-N. Chiu, J.-K. Wang, M.-T. Lin, E.-T. Wu, F. L. Lu, C.-I Chang, Y.-S. Chen, I.-S. Chiu, H.-C. Lue, and M.-H. Wu Aortic Valve Prolapse Associated With Outlet-Type Ventricular Septal Defect Ann. Thorac. Surg., April 1, 2005; 79(4): 1366 - 1371. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C. Rergkliang, V. Chittithavorn, A. Chetpaophan, and P. Vasinanukorn Surgery for Aortic Insufficiency Associated with Ventricular Septal Defect Asian Cardiovasc Thorac Ann, March 1, 2005; 13(1): 61 - 64. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
Y.-f. Cheung, C. S.W. Chiu, T.-c. Yung, and A. K.T. Chau Impact of preoperative aortic cusp prolapse on long-term outcome after surgical closure of subarterial ventricular septal defect Ann. Thorac. Surg., February 1, 2002; 73(2): 622 - 627. [Abstract] [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 |
