|
|
||||||||
Ann Thorac Surg 2005;79:1473-1479
© 2005 The Society of Thoracic Surgeons
a Department of Cardiovascular Medicine, Cleveland Clinic Foundation, Cleveland, OH
b Department of Thoracic and Cardiovascular Surgery, Cleveland Clinic Foundation, Cleveland, OH
c Department of Biostatistics, Cleveland Clinic Foundation, Cleveland, Ohio
Accepted for publication September 24, 2004.
* Address reprint requests to Dr Grimm, Department of Cardiovascular Medicine, Desk F15, Cleveland Clinic Foundation, 9500 Euclid Avenue, Cleveland, Ohio 44195 (E-mail: grimmr{at}ccf.org).
| Abstract |
|---|
|
|
|---|
METHODS: Perioperative echocardiograms on 132 consecutive patients (mean age 42 ± 12 years; 94% male), with bicuspid valves and isolated aortic regurgitation undergoing surgery at our institution were reviewed. Seventy-five patients (57%) underwent successful valve repair. Repair was attempted but unsuccessful for another 8 patients (6 intraoperatively and 2 before discharge).
RESULTS: Cusp prolapse was the most common primary mechanism of regurgitation (88 patients [67%]), with 81 patients having primarily eccentrically directed regurgitation. Echocardiographic examination of 72 (55%) had evidence of cusp thickening with 40 (30%) having cusp calcification. By multivariate analysis, an eccentric regurgitant jet direction (odds ratio = 14.3; 95% confidence interval [CI] = 3.4 to 59.6), lack of cusp thickening (odds ratio = 5.9 [1.7 to 20]), lack of cusp calcification (odds ratio = 4.2; [1.1 to 16.7]) and the absence of commissural thickening (odds ratio = 4.8 [1.3 to 16.7]) were independently associated with a greater likelihood of successful valve repair. Greater cusp thickening was the only factor associated with attempted but failed repair.
CONCLUSIONS: Successful repair of regurgitant bicuspid aortic valves was more feasible for those patients with eccentric regurgitant jets, those without cusp or commissural thickening or cusp calcification. Recognition of these features may enhance patient selection and improve procedural outcomes with aortic valve repair.
| Introduction |
|---|
|
|
|---|
Our knowledge regarding feasibility of repair for regurgitant BAVs is limited. Much of the decision making with regards to suitability for repair has been based on direct inspection of the valve at the time of surgery, a decision that would be preferable to make preoperatively. Echocardiography has the potential to fill this void. With the advent of major technical advances over the last decade, including marked improvements in image quality and the capability to perform multiplane transesophageal echocardiography (TEE), we now have an rapid, reliable, and accurate means to make a comprehensive assessment of valve morphology and function before surgery.
Using intraoperative echocardiography data from patients with regurgitant BAVs undergoing surgery, we sought to define the echocardiographic features of these valves, with the primary objective of identifying features that would help predict feasibility of successful valve repair. Improving our ability to predict suitability for repair should enhance patient selection, acute procedural success, and hopefully improve long-term outcomes.
| Patients and Methods |
|---|
|
|
|---|
3+) having surgery at our institution between January 1, 2000 and December 31, 2002 were identified from a search of our intraoperative TEE database. Those with aortic stenosis (aortic valve [AV] orifice area of < 2.0 cm2 or a mean transvalvular gradient of > 15 mm Hg), acute aortic dissection, focal sinus of Valsalva aneurysm, or a recent history (< 6 months) of infective endocarditis were excluded. This study was approved as part of our echocardiographic database registry by the institutional review board at the Cleveland Clinic Foundation. From our initial database search we identified 589 patients; 141 of these patients (24%) had BAVs. Nine patients were excluded (echocardiography images were unavailable for 3 patients, and 6 had requested valve replacement). Therefore, our study population consisted of 132 patients (mean age of 42 ± 12 years, with 125 male patients [94%]). The primary indications for surgery were dyspnea in 68 patients (52%), asymptomatic left ventricular (LV) dysfunction in 18 (14%), asymptomatic LV dilatation in 30 (23%), and aortic root dilatation in 15 (11%). There was documented evidence in the surgical notes that valve repair was considered in 118 patients (89%). It was assumed that repair was considered in the remaining patients, as it is a standard practice by all surgeons at the Cleveland Clinic Foundation to consider aortic valve repair as the preferred approach, if feasible.
Clinical and Operative Data
Clinical and operative data were obtained by review of patients' charts. Ten patients (8%) had prior cardiac surgery (7 had prior AV repair, 2 had previous ascending aorta replacement, and 1 mitral valve repair). Valve repair was attempted in 83 patients (62%), of whom 20 (24%) required a second round on cardiopulmonary bypass for residual AR of at least mild severity. Six of these attempted repairs were unsuccessful intraoperatively and required valve replacement. Before discharge from hospital, 2 patients developed significant recurrent AR (
2+) and underwent valve replacement. One patient had significant valvular obstruction post repair (mean transvalvular gradient of 41 mm Hg) and underwent a second "pump-run" with repositioning of commissuroplasty sutures, which reduced the mean gradient to an acceptable level (8 mm Hg). The final operative procedure ascribed to each patient was the last procedure performed by the day of discharge, with 75 patients (57%) having a valve repair (Fig 1). The clinical and echocardiographic features of the repair and replacement groups were compared to identify features that were associated with feasibility of successful valve repair.
|
|
The following features were recorded for each AV: primary mechanism of regurgitation; regurgitant jet direction; cusp fusion pattern; raphe thickness; commissural thickening; cusp thickening; and cusp calcification. The primary mechanism of regurgitation was reported as either cusp prolapse, restricted cusp motion (RCM) due to either cusp thickening, and retraction or relative RCM due to aortic root dilatation (Figs 2 and 3). All those with aortic root dilatation had echocardiographic evidence of intrinsic cusp thickening and therefore the latter two mechanisms were combined as RCM for analysis. Jet direction was defined as the direction of the largest regurgitant jet, either as eccentric (posterior or anterior) or central (Figs 2 and 3). To enhance reproducibility and clinical usefulness, commissural thickening, cusp thickening, and cusp calcification were dichotomized and reported as either absent or present. Commissural thickening was defined as the presence of increased echodensity at either commissure with associated commissural cusp separation (Fig 4). Cusp thickening was defined as the presence of any focal areas of increased echodensity (
2 mm) on either cusp with or without RCM [10]. Cusp calcification was defined as the presence of any area of typical increased echocardiographic brightness suggestive of calcification on either cusp (Fig 4), similar to the definition used by Abascal and colleagues [11] for mitral valve calcification.
|
|
Statistical Analysis
Statistical analysis was performed using SAS software (SAS Institute Inc., Cary, NC). Univariate comparisons between patients undergoing successful valve repair and replacement were made with
2 test, Student's t test, or Wilcoxon rank sum test as appropriate. For multivariate analysis, possible variables were preselected using bootstrap aggregation. In this method, a patient was selected at random with replacement from the original data set, and this selection was repeated the number of times that equals the number of patients in the original data set. Multivariate logistic regression analysis was applied to this data set with a variable retention p value of 0.05. Selected variables were recorded for a final model to identify variables that were associated with successful repair. The whole process was repeated 1000 times and variables with high occurrence frequencies were kept for a multivariate logistic regression analysis. Odds ratios for significant variables were calculated. Results were analyzed in three groups: all 132 patients; those 118 patients where there was documented evidence that valve repair was considered; and the 8 attempted but unsuccessful repairs. Aortic valve morphology was reassessed in a subgroup of 20 patients, with an acceptable variability (
value = 0.78 for commissural thickness, 0.84 for cusp thickness, and 0.88 for cusp calcification).
| Results |
|---|
|
|
|---|
|
Features Associated With Valve Repair
By univariate analysis, factors associated with a greater feasibility of having a successful valve repair were younger age, no prior cardiac surgery, smaller aortic dimensions, and absence of commissural thickening, cusp thickening or cusp calcification (Table 2). Repair was significantly more likely when the AR was due to cusp prolapse and when the primary AR jet direction was eccentric. Those who had other concomitant cardiac surgery were also significantly less likely to have a valve repair, especially if the surgery involved other valves. More patients requiring aortic root replacement underwent a composite valve and root replacement than a combined valve repair and root replacement procedure (23% vs 11%), although this did not reach statistical significance. Only 1 patient in the repair group had prior cardiac surgery (AV repair 5 years previous with a wedge resection of the anterior leaflet who had developed severe regurgitation due to recurrent prolapse of the conjoint leaflet) and underwent successful redo valve repair (anterior leaflet plication and commissuroplasty).
By multivariate analysis with adjustment for age, the following echocardiographic factors were independently associated with a greater likelihood of having a successful valve repair: an eccentric jet direction, absence of cusp thickness, lack of cusp calcification, and absence of commissural thickening (Table 3). We repeated the analysis for the subgroup (118 patients, 89%) where there was documented evidence that valve repair was considered and the same variables remained independently significant.
|
Predischarge Transthoracic Echocardiogram
Postoperative TTE was performed at a median of 4 days (interquartile range [IQR] = 2 to 7). Median peak and mean transvalvular gradients for the repaired valves were 21 mm Hg (IQR = 16 to 32) and 12 mm Hg (IQR = 9 to 18), respectively. Postrepair median EF was 50% (IQR = 45 to 55). No patients had residual AR of greater than 1+ in severity. Valve morphologic characteristics, including mechanism of regurgitation, cusp thickening, cusp calcification, commissural thickening, preoperative LV function, LV dimensions and aortic dimensions were not significantly associated with postoperative valve gradients or LV function.
| Comment |
|---|
|
|
|---|
The features identified as associated with a greater feasibility of successful repair in our series were the presence of eccentrically directed AR and the absence of cusp thickness, commissural thickness or cusp calcification. There were significant inter-relations between these factors, though in our final model, each emerged as independently significant. In BAVs the conjoint cusp typically has a longer free edge, which can prolapse and result in an eccentric regurgitant jet (most commonly a posteriorly directed jet due to prolapse of an anterior conjoint cusp) [13, 17]. Jet eccentricity can be considered as a surrogate for cusp prolapse and cusp tissue redundancy, especially in the absence of cusp thickening or calcification. Echocardiographically, it is often easier to appreciate AR jet direction than to convincingly show cusp prolapse, especially if the two-dimensional plane cuts along the long-axis of the conjoint cusp. This is particularly pertinent with transthoracic studies, where detailed aortic valve anatomy is often difficult to elucidate, though jet direction is usually obvious. All the features identified as significant in our series are age related, with those features associated with less likelihood of valve repair being more common with increasing age. In our multivariate model, these features remained significant despite including the interaction between cusp thickening and age. With increasing age, chronic progressive cusp thickening and calcification results in restricted retracted cusps, which are less suited to repair. Our findings corroborate previous reports where this "deficient cusp tissue" pathology was less amenable to repair [13, 16]. It also appears to suggest that the timing of surgery may be important when considering valve repair. Earlier intervention before the development of these age related changes may enhance repair potential.
Our series of patients is similar in age profile, gender, and primary mechanism of regurgitation to other reported series of regurgitant BAVs [4, 18, 19]. We specifically excluded those with acute infective endocarditis (reported incidence of up to 30% of regurgitant BAVs [18]), as surgical intervention on these valves not only depends on the original appearance of the BAV but also on the destruction caused by the infection. Our reported repair rate for regurgitant noninfected BAVs (57%) is lower than a previous reported series from our group (76%), although that series was restricted only to patients with cusp prolapse [12]. The factors identified by our study need further prospective evaluation to confirm the true proportion of regurgitant BAVs that may be suitable for valve repair.
Cusp thickness was the only factor identified in the small group of patients (n = 8; 10% of repairs) where repair was attempted but unsuccessful. The small size of this group limits our capacity to identify other factors, whereas the finding that similar features are important in early repair failure because those associated with repair feasibility supports the validity of our results.
Limitations
This retrospective observational study is subject to all the limitations attributed to such a study. Our assumption that all patients were considered for valve repair appears valid because a repeat analysis, confined to those where there was documented evidence that repair was considered, found the same findings. It is important to emphasize that in our overall analysis, the factors we identified are factors associated with the likelihood of having a repair or not (although our analysis was limited to comparing those who had a successful repair or not). The subgroup of 8 patients who had attempted but unsuccessful repairs were the only patients where factors associated with the success or failure of a repair can be accurately analyzed. Patients in this study underwent valve repair by surgeons with recognized expertise in valve repair, hence these results may not be generalizable. However, preoperative detection of patients that may be suitable for repair may allow appropriate referral to surgeons with the required expertise. Despite many of the reported morphologic features being subjective in nature (cusp thickening, commissural thickening, or cusp calcification), our predefined criteria appear to have satisfactory reproducibility.
This study is a feasibility study aimed at identifying which echocardiographic features of regurgitant BAVs are associated with a greater likelihood of achieving successful valve repair, and it is not aimed at reporting long-term outcome data with this procedure. The intermediate-term durability of repair has been reported by a number of groups, with freedom from reoperation at 5 years of between 87% and 91% [12, 19, 20]. Comprehensive long-term follow-up of this and other series of patients who have undergone aortic valve repair is necessary and currently ongoing. Data on recurrent AR is more limited, reflecting the lack of routine echocardiographic follow-up in most reported series to date. Davierwala and colleagues [19] report a freedom from moderate or severe AR (> 2+) of 79% for valve repair compared with 94% for valve replacement. Currently the intermediate-term outcomes with valve replacements are excellent (> 95% freedom from reoperation at 5 years) [19]. Despite valve replacement being the current standard surgical approach, it is not without long-term risks. Complications due to anticoagulation with prosthetic valves and the predictable degeneration of heterografts and homografts become increasingly significant at long-term follow-up (>10 years). We expect that with greater experience, knowledge, and understanding of AV function and repair techniques, the intermediate-term outcomes with repair should improve further. The gold standard for valve repair will have to be an average performance equal to or better than a heterograft, with the potential for real long-term function.
Conclusions
Recognizing the fact that none of the presently available options for aortic valve replacement are ideal for these typically younger patients with BAVs and isolated AR, valve repair is a potentially important alternative. Until now only limited information has been available on which to base management decisions as regards feasibility and suitability for aortic valve repair. Our study identifies a number of features that may help improve preoperative assessment of valve reparability, guide and optimize surgical management, and ultimately may have a possible influence on long-term outcome after aortic valve repair. Adequate assessment of these valves requires TEE imaging and we suggest that all patients with regurgitant BAVs undergoing surgery should be considered for TEE to assess for potential reparability.
|
| References |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
J.-B. le Polain de Waroux, A.-C. Pouleur, A. Robert, A. Pasquet, B. L. Gerber, P. Noirhomme, G. El Khoury, and J.-L. J. Vanoverschelde Mechanisms of Recurrent Aortic Regurgitation After Aortic Valve Repair: Predictive Value of Intraoperative Transesophageal Echocardiography J. Am. Coll. Cardiol. Img., August 1, 2009; 2(8): 931 - 939. [Abstract] [Full Text] [PDF] |
||||
![]() |
G. B. Pettersson, A. C. Crucean, R. Savage, C. M. Halley, R. A. Grimm, L. G. Svensson, S. Naficy, A. M. Gillinov, J. Feng, and E. H. Blackstone Toward predictable repair of regurgitant aortic valves a systematic morphology-directed approach to bicommissural repair. J. Am. Coll. Cardiol., July 1, 2008; 52(1): 40 - 49. [Abstract] [Full Text] [PDF] |
||||
![]() |
J. S. Borer and L. N. Girardi Repair of the congenitally bicuspid regurgitant aortic valve a strategic advance. J. Am. Coll. Cardiol., July 1, 2008; 52(1): 50 - 51. [Full Text] [PDF] |
||||
![]() |
J.-B. le Polain de Waroux, A.-C. Pouleur, C. Goffinet, D. Vancraeynest, M. Van Dyck, A. Robert, B. L. Gerber, A. Pasquet, G. E. Khoury, and J.-L. J. Vanoverschelde Functional Anatomy of Aortic Regurgitation: Accuracy, Prediction of Surgical Repairability, and Outcome Implications of Transesophageal Echocardiography Circulation, September 11, 2007; 116(11_suppl): I-264 - I-269. [Abstract] [Full Text] [PDF] |
||||
![]() |
H. Dave and R. Pretre Pericardial patch reconstruction of the congenitally diseased aortic valve MMCTS, May 7, 2007; 2007(0507): 1354. [Abstract] [Full Text] [PDF] |
||||
![]() |
H.-H. Sievers and C. Schmidtke A classification system for the bicuspid aortic valve from 304 surgical specimens J. Thorac. Cardiovasc. Surg., May 1, 2007; 133(5): 1226 - 1233. [Abstract] [Full Text] [PDF] |
||||
![]() |
G. Maurer Aortic regurgitation. Heart, July 1, 2006; 92(7): 994 - 1000. [Full Text] [PDF] |
||||
![]() |
S. H. Rahimtoola The Year in Valvular Heart Disease J. Am. Coll. Cardiol., January 17, 2006; 47(2): 427 - 439. [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 |