|
|
||||||||
a Cardiac Surgery Unit, Magna Graecia University of Catanzaro, Catanzaro, Italy
b Cardiology Unit, Magna Graecia University of Catanzaro, Catanzaro, Italy
Accepted for publication May 18, 2009.
* Address correspondence to Dr Rubino, Cardiac Surgery Unit-Magna Graecia University of Catanzaro, Viale Europa, Località Germaneto, Catanzaro, 88100, Italy (Email: cchumg{at}hotmail.it).
| Abstract |
|---|
|
|
|---|
Methods: Thirty patients with functional mitral regurgitation were prospectively randomized to RMA alone or cardiac restraint with the CorCap Cardiac Support Device (Acorn Cardiovascular Inc, St. Paul, MN) and RMA. Clinical, echocardiographic, New York Heart Association (NYHA) functional class, Short Form 36-Item Health Survey (SF-36) quality of life scores, and N-terminal pro-B-type natriuretic peptide (NT-proBNP) results were analyzed.
Results: No hospital deaths or device-related complications occurred. The two groups had comparable morbidity (p = 0.34). Echocardiography showed a trend towards a slightly better functional improvement during follow-up in CorCap plus RMA patients (between groups, p = 0.001). Both groups showed improved results for SF-36, NYHA, and NT-pro.BNP; however, CorCap plus RMA patients had significantly better SF-36 at discharge (p = 0.003), postoperative NYHA (p = 0.05), and NT-pro.BNP (p = 0.001). Survival (p = 0.46), freedom from CHF (p = 0.23), and rehospitalization (p = 0.28) were comparable. Patients in whom CHF developed after postoperative day 1 had higher NT-pro.BNP values (p = 0.001 at all time-points).
Conclusions: Adjunctive application of CorCap with RMA correlated with better NT-pro.BNP at short-term follow-up together with slightly improved echocardiographic and functional results. This deserves further evaluation at midterm and long-term follow-up. Reduction of NT-pro.BNP at follow-up may be suggested as a prognostic index.
| Introduction |
|---|
|
|
|---|
Cardiac restraint devices, such as the CorCap Cardiac Support Device (Acorn Cardiovascular Inc, St Paul, MN), have been regarded as means to limit or to prevent chronic cardiac dilation and the ensuing progression of CHF [7]. In particular, experimental models showed that CorCap could address the maladaptive mechanisms associated with increased wall stress [8, 9]. Such promising results drew attention to CorCap as a novel strategy to improve cardiac function [8, 9].
Accordingly clinical trials validated the feasibility of the technique, showing improvement of clinical symptoms, as well as hemodynamics and echocardiographic results [10–17]. However, the potential benefit of combining passive cardiac restraint devices with mitral valve (MV) procedures is still questionable [18]. Therefore, the aim of this study was to discern the early differences in neurohormonal release and echocardiographic findings in a subset of patients with CHF secondary to dilated cardiomyopathy and FMR, randomized to receive either CorCap plus restrictive mitral annuloplasty (RMA) or RMA alone.
| Material and Methods |
|---|
|
|
|---|
Patients and Inclusion Criteria
Between April 2007 and April 2008, 30 consecutive patients with FMR, eligible for RMA, were prospectively randomized by lottery to CorCap plus RMA (CorCap group) or RMA alone (RMA group). Inclusion criteria were New York Heart Association (NYHA) functional class III and IV, age between 18 and 80 years, FMR grade 3+ or 4+ due to severe annular dilation and dilated cardiomyopathy (Carpentier type I) at preoperative echocardiography (semiquantitatively from color-flow Doppler), left ventricular ejection fraction (LVEF) of 0.35 or less, and LV dilation, which was defined as LV end-diastolic diameter (LVEDD) of 6.5 cm or more despite optimal medical management including diuretics, angiotensin-converting enzyme inhibitors, and β-blockers. Tricuspid regurgitation 2+ or more at transthoracic echocardiography indicated the need for tricuspid annuloplasty. Mean arterial pressure and heart rate were monitored preoperatively (at hospital admission), at discharge, and at the end of follow-up. Follow-up was closed on October 31, 2008, and was 100% completed.
Patients requiring a concomitant operation, other than tricuspid annuloplasty, were excluded from the study.
Echocardiography
All echocardiograms were performed by either of 2 cardiologists, using a VIVID 7 Pro ultrasound machine (GE Technologies, Milwaukee, WI). All patients had transthoracic echocardiographic assessment preoperatively, at hospital discharge, and at midterm follow-up, consisting of standard examination, including grading of FMR and transmitral mean gradient semiquantitatively from color-flow Doppler, LVEDD, LV end-systolic diameter (LVESD), LVEF, and sphericity index (calculated as the long/short axis ratio). Recurrence of CIMR was defined as 3+/4+ or higher grade MR at any time postoperatively at semiquantitative color-flow Doppler analysis.
Intraoperative transesophageal echocardiograms (TEE) before the establishment of cardiopulmonary bypass (CPB) and after weaning from CPB ensured adequate CorCap fitting, defined as LVEDD reduction of about 10% of preoperative values. LVEF and valve function were also assessed intraoperatively. MV repair was considered successful at intraoperative TEE if there was no residual FMR after CPB discontinuation with an achieved adequate preload (central venous pressure between 10 and 15 mm Hg).
Biochemical Assays
Blood samples for assessment of N-terminal prohormone brain natriuretic peptide (NT-pro.BNP) levels were obtained preoperatively (before anesthetic induction), on postoperative day 1, at discharge, and at the end of follow-up. Samples were analyzed by an electrochemiluminescence immunoassay (Elecsys proBNP) using an Elecsys 2010 analyzer (Roche Diagnostics, Mannheim, Germany).
CHF Evaluation
NYHA and quality of life according to the Short Form 36-Item (SF-36) Health Survey were determined preoperatively, at discharge, and at the end of follow-up in the outpatient clinic. As a surrogate outcome of heart failure, the need and dose of furosemide were assessed preoperatively, on postoperative day 1, at discharge, and at the end of follow-up in the outpatient clinic.
Surgical Procedure
The RMA operation was performed using standard operative techniques, including CPB and undersized mitral annuloplasty. The patients were placed under general anesthesia, and the same surgeon (A. R.) performed all operations through a median sternotomy. The MV was exposed through a longitudinal atriotomy along the Waterston groove in all patients.
Ring size (Carpentier-Edwards Physio ring; Edwards Lifesciences, Irvine, CA) was determined after careful measurements of the height of the anterior leaflet and the intertrigonal distance, and then downsizing by two sizes (eg, size 26 when measuring 30). CPB was standardized: the ascending aorta was always cannulated and venous return always accomplished through a double caval cannulation. A Dideco (Mirandola-Modena, Italy) tubing set, which included a 40-µm filter, a Stockert roller pump (Stockert Instrumente, Munich, Germany), and a hollow-fiber membrane oxygenator (Dideco D903 Avant, Mirandola Modena, Italy) were used. Myocardial protection was always achieved with intermittent antegrade and retrograde hyperkalemic blood cardioplegia.
Perioperative need of inotropic drugs was recorded. We have previously shown that phosphodiesterase inhibitors can reduce troponin I release and the need for further inotropic support [19]; therefore, we started enoximone (5 µg/kg/min) immediately after aortic cross-clamp removal. Further increase of inotropic support was recorded and defined as low dose when 5 µg/kg/min or less of enoximone was given, medium dose when 6 to 10 µg/kg/min of enoximone was given or 5 to 10 µg/kg/min of dobutamine was added, and high dose when > 10 µg/kg/min of enoximone or dobutamine was given, or when any dose of epinephrine was added [19].
The anticoagulation protocol consisted of 4000 IU nadroparin daily, initiated 8 hours postoperatively. This was followed by warfarin therapy starting on postoperative day 2 (targeted to an INR between 2.5 and 3.5) and continuing until postoperative month 3, or life-long in patients demonstrating longstanding persistent atrial fibrillation.
Study End Points
Primary end points for the study included echocardiographic indexes (LVEF, LVEDD, LVESD, FMR, and LV sphericity index) and NT-pro.BNP values. Secondary end points were hospital death and morbidity, change in NYHA functional class, quality of life assessment, follow-up survival, and freedom from CHF and rehospitalization. Device-related complications were also recorded.
Statistical Analysis
The interaction between echocardiographic and NT-pro.BNP values, and within the three study time points (preoperative, at discharge, and after follow-up) in 15 patients for each study arm gave a power (1-β error probability) of 84% with an
-error probability of 0.05. Cumulative survival curves for all events were constructed according to the Kaplan-Meier method, and differences were tested by the log-rank test. Comparisons of changes from baseline to follow-up were evaluated with t tests,
2 statistics, two-way analysis of variance (ANOVA), and with general linear model for repeated measures, as appropriate. All differences were considered significant at the value of p < 0.05 (two-sided).
| Results |
|---|
|
|
|---|
|
|
|
|
Accordingly, the sphericity index improved during the follow-up in both groups (CorCap, p = 0.001; RMA, p = 0.013); however, the sphericity index values were similar at single time points and by between-group analysis. EF was similar preoperatively between the two groups but was higher in the CorCap group after hospital discharge (p = 0.001). Patients in both groups showed improved EF throughout the follow-up (p = 0.001), but patients randomized to CorCap showed better improvement of EF compared with RMA (p = 0.001).
Semiquantitative grading of recurrent MR showed the CorCap and RMA patients were not statistically different from each other in terms of single-time comparison or in terms of within- and between-group relationships (Table 3). Moreover, patients were similar at follow-up either for the recurrence rate of MR or for mean transmitral gradients (Table 4).
|
|
When biochemical markers of heart failure were compared, NT-proBNP levels were similar preoperatively and at postoperative day 1 (Table 6). However, both groups showed reduction of NT-proBNP levels during course of the study (CorCap group, p = 0.001; RMA group, p = 0.001). Accordingly, patients in the CorCap group showed lower NT-proBNP levels by between-group analysis (p = 0.005).
|
| Comment |
|---|
|
|
|---|
RMA has been successfully used for more than a decade and has been accepted as an effective means for immediate improvement in valve function [21]. Despite the enthusiasm for a low perioperative mortality rate, a positive effect on survival, decreased CHF and functional NYHA class, a number of studies have clearly demonstrated that RMA is subject to variable rates of recurrence [22, 23]. In a cohort of patients mainly representing nonischemic CHF and severe LV dysfunction, the Acorn trial demonstrated that MV operation was safe, carried a low operative mortality rate, and was associated with a remarkable degree of beneficial LV reverse remodeling, This was manifested by reductions of the ventricular volumes, improvements in EF and sphericity index, and a reduction in LV mass [16].
Our report similarly showed that RMA with downsizing by two ring sizes can successfully restrict the mitral annulus and provide early signs of LV reverse remodeling, as manifested by a decrease in LVEDD and LVESD, an improvement in EF and sphericity index, along with evidence of improvement in quality of life. On the other hand, a recent study raised concern that the higher mean transmitral gradients after RMA may act to promote "iatrogenic" mild mitral stenosis, preserving the LV from progressive dilation [24]. Accordingly our data showed some sort of mitral stenosis at echocardiographic Doppler analysis (Table 4).
Because RMA addresses LV dilation specifically at the level of the annulus, there is limited beneficial effect at the midventricular and apical levels. This allows ongoing ventricular dilation and remodeling to occur, with a net result of FMR recurrence during follow-up [22, 23]. In fact, the Acorn trial demonstrated that recurrent FMR after an initial repair with ring annuloplasty paralleled the increase of LV volumes and of the sphericity index, suggesting that ongoing ventricular remodeling can overwhelm the effects of the annuloplasty ring. Therefore, a number of complementary surgical strategies have been suggested, including ventricular restraint devices and surgical ventricular restoration [25].
Currently, there are no specific therapies designed to address progressive LV dilation, which is one of the strongest predictors of death for patients with CHF [26]. Passive containment by the CorCap device was first reported in an animal model of chronic dilated cardiomyopathy [7–9]. These studies demonstrated that CorCap-treated hearts experienced significant decreases in volume and improvement in function, all consistent with reverse remodeling. A reversal of remodeling on a cellular and molecular level was further demonstrated [7–9].
The Acorn trial also demonstrated a lack of constrictive physiology, despite significantly greater reductions in LV volume and greater improvement in sphericity when a CorCap was added to MV operations [16]. Subsequent studies with the same cohort of patients confirmed that CorCap is a useful adjunct to MV operations in the setting of dilated cardiomyopathy [17]. Accordingly, our data seem to suggest that CorCap added to MV interventions can promote ventricular reverse remodeling by effecting reduction of LVEDD and LVESD (Table 3).
Mechanistic studies in animal models have provided evidence that the CorCap leads to fundamental changes in the biology of the myocytes, including reversal of fetal gene expression, increased adrenergic sensitivity, decreased interstitial fibrosis, increased capillary density, and reduced myocyte hypertrophy [26–28]. Whether these benefits accrue to patients treated with CorCap and RMA remains to be seen.
Our echocardiographic results could only suggest a trend towards ventricular reverse remodeling. It could be speculated that ventricular containment might not have substantially altered the natural history of heart failure progression. However, the combination of CorCap with RMA resulted in a clinical benefit for patients, as represented by reduced NYHA at follow-up and improved SF-36 results in the group that received CorCap and RMA. Accordingly, the CorCap patients had less need for furosemide at follow-up.
Evaluations of plasma BNP and NT-proBNP levels are being increasingly incorporated into clinical practice, especially for assessment, prognostic stratification, and management of heart failure [29]. A relevant finding of the present study was that the higher the NT-proBNP preoperative level, the greater the incidence of follow-up CHF and ventricular remodeling. On the other hand, the greater the postoperative decrease in NT-proBNP leakage, the more extensive LV reverse remodeling occurred. Such evidence underscores an important biologic role for NT-proBNP in the remodeling process.
Ventricular wall stress with accompanying diastolic fiber stretch promotes overexpression of NT-proBNP by angiotensin II signaling [30], in association with stress-dependent, angiotensin-mediated myocardial hypertrophy and interstitial fibrosis [31]. Notably, our data showed that CorCap and RMA correlated with a more significant reduction in NT-proBNP postoperatively and at follow-up, than RMA alone. These data confirmed results from experimental as well as clinical studies [31].
According to the role of NT-proBNP in matrix remodeling, a strong reduction of NT-proBNP levels after CorCap with RMA could identify a category of patients in which a lower degree of fibrosis may develop. Similarly, a persistently high NT-proBNP level in CorCap patients may anticipate absence of meaningful LV remodeling with poor follow-up prognosis. This direct relationship between postoperative LV reverse remodeling and NT-proBNP levels suggests that neurohormonal assays may be an adjunct to perioperative echocardiography for prognostic stratification of patients undergoing CorCap and RMA for CHF. However, further studies on this topic are necessary to test this hypothesis.
Finally, we emphasize that no CorCap-related complications were registered, and that the procedure was not time-intensive. These results suggest that the CorCap can be a useful tool in an integrated surgical treatment of CHF and FMR.
The main limitation of this study is that the number of enrolled patients was probably not adequate to obtain a sufficient power to discriminate differences between patients receiving and not receiving the CorCap device. Moreover, the short follow-up enables discussion only of early results after CorCap implantation. Further studies enrolling a larger number of patients and incorporating a longer follow-up will be required to assess the effect of CorCap on the progression of CHF. These limitations, however, are the results of the single-center study design, which consequently insures uniformity of perioperative patient management and follow-up throughout the study.
Likewise, the strength of our study reflects incorporation of a prospective design with sequential echocardiographic assessments and neurohormonal evaluations, and a 100% completed follow-up.
In conclusion, results of our study support the hypothesis that CorCap may be a useful adjunct to RMA in CHF, allowing better functional recovery, quality of life, and neurohormonal response. In particular, postoperative reduction of NT-proBNP may be suggested as a prognostic index in this cohort of patients.
| References |
|---|
|
|
|---|
Related Article
This article has been cited by other articles:
![]() |
M. Vaynblat and M. Chiavarelli Ventricular Constraint in Heart Failure Ann. Thorac. Surg., April 1, 2010; 89(4): 1344 - 1345. [Full Text] [PDF] |
||||
![]() |
A. S. Rubino, F. Onorati, and A. Renzulli Reply Ann. Thorac. Surg., April 1, 2010; 89(4): 1345 - 1345. [Full Text] [PDF] |
||||
![]() |
J. Raman Invited Commentary Ann. Thorac. Surg., September 1, 2009; 88(3): 726 - 726. [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 |